Home

USARSim Manual

1. Where Type string string is the sensor effecter s type name Name string Location Specifies how an item is mounted on the x y z Orientation x y z robot The string after Name is the item s Mount string name and the string after Mount is the item s mounting base which is also another item The x y z after Location and Orientation are the item s relative location and orientation to the mounting base If more than one sensor effecter is the same type multiple Name string Location segments will appear in the GEO message Example GEO Type Camera Name Camera Location 0 0820 0 0002 0 0613 Orientation 0 0000 0 0000 0 0000 Mount CameraTilt For a mission package the geometry message is expressed in the following format to tell us how the mission package and its elements are installed together to the robot GEO Type MisPkg Name string Link int ParentLink int Location x y z Orientation x y z Link int ParentLink int Location x y z Orientation x y z Where Type MisPkg The Type parameter is hard coded and will always be MisPkg for a mission package GEO message Name string string is the mission package s name Link int int is the link number of the described link 60 ParentLink int in is the parent link number of the described link If the described link is mounted directly on the robot s chassis
2. string is the name of the camera currently attached to viewportl If viewportl has been disabled this parameter will be Disabled If viewport has been attached to a non existent camera this parameter will be None Status string string is the status of viewportl after the SET command has been issued The status will be OK when the camera attached to viewportl was successfully changed Otherwise the status will be Failed Viewport2 string string is the name of the camera currently attached to viewport2 If viewport2 has been disabled this parameter will be Disabled If viewport2 has been attached to a non existent camera this parameter will be 65 Status string Viewport3 string Status string Viewport4 string Status string None string is the status of viewport2 after the SET command has been issued The status will be OK when the camera attached to viewport2 was successfully changed Otherwise the status will be Failed string is the name of the camera currently attached to viewport3 If viewport3 has been disabled this parameter will be Disabled If viewport3 has been attached to a non existent camera this parameter will be None string is the status of viewport3 after the SET command has been issued The status will be OK when the camera attached to viewport3 was successfully changed Oth
3. Example SET Type Camera Name Camera FOV 1 will set the field of view of camera Camera to 1 radian SET Type Camera Name Camera FOV 0 will set the field of view of camera Camera to its default field of view e Control a Mission Package A mission package is constructed of a series of connected elements Of course we can control the joints one by one to set the mission package s pose Here we provide another command to directly set the package s terminal pose and let USARSim control every element s joint for us If a camera is mounted on a pan tilt mission package we can use this command to control the camera s pose Using mission package control commands we can have multiple cameras and control them separately The command s format is MISPGK Name string Link int Value float Order int Link int Value float Order int Where Name string Link int Value float Order int string is the mission package s name int is the link number that will be moved using the next parameters The float parameter is the value used to move the link What this parameter describes depends on the order given If the order is 0 float is the absolute angle in radians for a revolute joint or the distance in meters for a prismatic joint If the order is 1 float is the velocity in m s If the order is 2 float is the torque int is optio
4. REFPOSE Seq 0 C Hand EClroretom Bic Yi L_Foot R_Foot Figure 517 Skeletal bones name For more details about skeletal mesh please visit UDN AnimBrowserReference http udn epicgames com Two AnimBrowserReference UDN UWSKkelAnim2 http udn epicgames com Two SkelAnim2 After you set the actions the victim will not move immediately In Unreal Editor everything is static To let them to be active you need to play the map As we know there is a bug in the Unreal engine Some meshes may play their default animations when your viewpoint is far away from the victim NOTE There is hierarchical relationship in the skeletal system Changing one scale value may affect other segments under it For example hips will affect thighs shins and feet 14 2 Build your sensor Before you build your sensors you need to understand Unreal Script and the client server architecture of the Unreal engine The following resources may be helpful to you 146 UDN UnrealScriptReference http udn epicgames com Two UnrealScriptReference UnrealWiki UnrealScript Topics http wiki beyondunreal com wiki UnrealScript Unreal Networking Architecture http unreal epicgames com Network htm 14 2 1 Overview In USARSim all sensors are inherited from the Sensor class The Sensor class defines the interfaces that the robot model can interact with We use a hierarchical architecture to build the sensors The hierarchy ch
5. SET Type RFID Name RFID Opcode Read Params RFIDTagID Where RFIDTagID is the ID of the RFIDTag you want to read RES Time Type RFID Name RFID Status OK ID RFIDTagID Mem MemoryContent Failure RES Time Type RFID Name RFID Status Failed Success For example failure can happen when the RFID tag is out of the sensor range 10 9 6 Writing RFID Tags Memory Using the same RFIDSensor you can write the RFIDTag memory using this command SET Type RFID Name RFID Opcode Write Params RFIDTagID MemoryContent Where RFIDTagID is the ID that identifies the RFIDTag you want to write MemoryContent is the string you want to write in the RFIDTag Success RES Time Type RFID Name RFID Status OK Failure RES Time Type RFID Name RFID Status Failed 10 9 7 Erasing RFID Tags Memory To erase the RFIDTag memory you can both write a 0 string or more easily SET Type RFID Name RFID Opcode Write Params RFIDTagID Where RFIDTagID is the ID that identifies the RFIDTag you want to erase Success RES Time Type RFID Name RFID Status OK Failure RES Time Type RFID Name RFID Status Failed 93 10 10 Victim and False Positive Sensor 10 10 1 How the sensor works The Victim and False Positive Sensor simulates a victim location sensor The sensor s operation is very similar to that of the Range Scanner Indeed the victim sensor sends out a number of
6. command that allows the user to set the current INS estimation of the position and orientation NOTE This will not set the sensor estimation back to ground truth Also if the Drifting mode is active this command will reset the drifting to 0 POSE X Y Z 0 Y OK successfully set the robot s estimation of pose Failed didn t set the robot s estimate of pose It may be caused by an invalid command or the absence of commas between arguments 9 5 2 How to configure it The Gaussian random number generator uses the Box Muller method and is based on a mean and sigma This function is included in a class called USARUtils uc This function requires persistent variables and therefore needs to be implemented as an object and not as static functions The sensor works in two ways In its simplest form it does not drift over time and its accuracy is determined by the sigma parameter only It is however also possible to set the sensor so that it drifts over time thus resembling error dynamics due to double integration observed in real sensors When drifting is enabled the Precision parameter characterizes the drift rate Please note that when drifting components drift in possibly different directions and at different rates The initialization of the USARUtils object is in sensor uc so that all sensors can utilize this function Therefore all sensors can use a Sigma in the USARBot ini file to adjust the distribution of noise
7. string is the robot s name Dimensions x y z x defines the robot s length y defines the robot s width and z describes the robot s height Please note that these values are in meters COG x y z x y and z identify the position of the center of gravity in meters calculated from the chassis origin Example GEO Type NauticVehicle Name Submarine Dimensions 6 5364 1 1428 1 9929 COG 0 0000 0 0000 0 0000 For an aerial vehicle the geometry information message looks like still to be expanded GEO Type AerialVehicle Name string Dimensions x y z COG x y Z Where Type AerialVehicle The Type parameter is hard coded and will always be AerialVehicle for an aerial vehicle GEO message Name string string is the robot s name 59 Dimensions x y z x defines the robot s length y defines the robot s width and z describes the robot s height Please note that these values are in meters COG x y z x y and z identify the position of the center of gravity in meters calculated from the chassis origin Example GEO Type AerialVehicle Name AirRobot Dimensions 9 0082 6 6897 2 6128 COG 0 0000 0 0000 0 0000 For sensors or effecters the geometry message looks like GEO Type string Name string Location x y z Orientation x y z Mount string Name string Location x y z Orientation x y z Mount string
8. MaxSpeed float MaxTorque float MinRange float MaxRange float Link int JointType string MaxSpeed float MaxTorque float MinRange float MaxRange float Where Type MisPkg Name string Link int JointType string MaxSpeed float MaxTorque float MinRange float MaxRange float The Type parameter is hard coded and will always be misPkg for a mission package CONF message string is a mission package s name int is the mission package s link number that will be described string can either be Revolute or Prismatic as determined by the type of joint being described float describes the joint s maximum speed in rad s float describes the joint s maximum torque For a revolute joint float is the minimum absolute angle that the joint can rotate to For a prismatic joint float is the minimum distance that the joint can move to For a revolute joint float is the maximum absolute angle that the joint can rotate to For a prismatic joint float is the maximum distance that the joint can move to NOTE For revolute joints if the MinRange parameter is greater than the MaxRange parameter the joint does not have any constraints Example CONF Type MisPkg Name CameraPantilt Link 1 JointType Revolute MaxSpeed 0 17 MaxTorque 20 00 MinRange 1 MaxRange 0 Link 2 JointType Revolute MaxSpeed 0
9. ROLL NORMAL VELOCITY p VELOCITY w yim t 2 AXIS YAW AZ WELOCITY ir Figure 18 SAE J670 Vehicle Coordinate System In Unreal a rotator is represented by the following structure Every element is an integer in Unreal Units struct Rotator var config int Pitch Yaw Roll However in the application interface we use a vector to describe a rotator The X Y and Z element represent the rotation angle around the corresponding axis They are all floating point values expressed in radians 6 2 Units and scale The unit used in Unreal is called an UU Unreal Unit Unreal Engine uses it to represent both length and angle The exceptions in Unreal Engine are 1 it uses degrees instead of UU to count FOV Field Of View 2 in trigonometric functions it uses radians The unit conversion is summarized below 250 UU 1 m Please use the function C_MeterToUU to convert from UU to meters 32768 UU 3 1415 radian 180 degree 0 5 circle In the application interface we use SI units that are built upon the modern metric system The base SI units along with the symbols used for abbreviations are listed in Table 1 By default all the data sent out from USARSim is represented as a floating point number that has 4 digits after the decimal point Table 1 SI base units SI base unit 47 amount of substance Mole mol Luminous intensity NOTE In your application all the data you get from USARSim is i
10. USARBot INS HiddenSensor true bWithTimeStamp False Weight 0 1 ScanInterval 0 2 Drifting false Precision 1000 Sigma 0 1 Where HiddenSensor Boolean value is used to indicate whether the sensor will be visually shown in the simulator Setting it to true will hide the sensor We recommend setting it to true if it s not necessary to show the sensor When you want to confirm if the sensor is placed in the correct position and has the correct direction you can temporarily set it to false bWithTimeStamp Indicates whether the timestamp in the Unreal Engine will be associated with the sensor data Weight The weight of the sensor in kg 87 ScanInterval The time interval in seconds between pose estimates Mean Sets the mean used by the Gaussian random number generator used to produce noise in the sensor Sigma Sets the standard deviation used by the Gaussian random number generator used to produce noise in the sensor RightTire The right wheel that will be used in pose estimating If it is a null string the right most wheel will be used Drifting Boolean value If true the sensor drifts if false it does not default is false Precision Numeric value indicating the rate of drifting The higher the value the less the drifting default is 1000 This parameter has effect only when Drifting is true 10 7 Encoder Sensor 10 7 1 How the sensor works The sensor measures a part s spin angle around the sensor s axis The returned
11. USARBot kenaf to represent this robot In summary the Kenaf has Two full body tracks to avoid stacking on a pole in step field Four flippers each of which can be controlled independently Powered by 6 x 50W brush less DC motors The Kenaf s specification is as follows Dimensions Minimum Length 575 mm Maximum Length 937 mm include flipper length Width 429 mm Height 259 mm without sensors and cameras Weight approx 20 kg In our simulation it is equipped with One PTZ camera One Bird s eye view camera One 2D Hokuyo laser scanner Six encoder sensors 127 Sensors and effectors can be attached to the Kenaf model in the same way as to any other robot as for example a P2AT To control the arm in USARSim we use the following commands Flipper control MULTIDRIVE FRFlipper x FLFlipper y RRFlipper z RLFlipper r where x y z r are the flipper angles in radians The four parameters control the front left front right rear left and rear right track separately For example to stow the tracks to lift or lower the chassis we use MULTIDRIVE FRFlipper 1 5 FLFlipper 1 5 RRFlipper 1 5 RLFlipper 1 5 or MULTIDRIVE FRFlipper 1 5 FLFlipper 1 5 RRFlipper 1 5 RLFlipper 1 5 Camera control MULTIDRIVE kenafCameraPan x kenafCameraTilt y where x y are the pan tilt angles in radians For example to pan the camera to left or right side 1 0 radians MULTIDRIVE
12. and Slower buttons speed up or slow down the robot s moving speed The Stop button stops the robot The buttons on the top Turn More Straight and Turn Less are only effective for Ackerman steered vehicles Turning a skid drive vehicle with separately powered wheels can be done by clicking the left or right arrow This will send opposite velocities to each wheel Camera group This group controls the robot s camera Up and down arrow buttons tilt the camera up and down 10 degrees Left and right arrow buttons pan the camera to left and right side 10 degrees The Zoom In and Zoom Out buttons zoom in and zoom out the camera separately Light group The Lights On Lights Off button turns the headlights on or off The Trace On Trace Off will enable or disable leaving a green trace of where the robot has been The trace will persist until you restart the server 34 Control Interface Wheels gt Turn Tum More Straight Less A ll gt ll WM Fast stp stone A Il Il SA KA Camera lt Lights On Update FPS 3 14077 Width 320 Height 240 Hide UT Trace On Time 244 23 Robot State E Chassis Rotation Riad Pitch 0 00 Roll 0 00 Yaw 0 00 Location m X 73 24 Y 54 98 Z 0 18 Velocity m s X 0 00 Rotation Rad Pitch 0 00 Roll 0 00 Yaw 0 00 Location m X 73 24 Y 54 98 Z 0 18 Wel
13. e Dimensions 121 cm x 12 2 cm x 14 5 cm e Maximum translate speed 370 mm s e Maximum step height 483 mm 12 15 2 Configure it It s the same as the P2AT 12 16 Kurt2D 12 16 1 Introduction The Kurt2D is a high speed indoor robot In USARSim we use classname USARBot Kurt2D to represent this robot 118 a Real Kurt2D b Simulated Kurt2D Figure 395 Kurt2D Robot In summary the Kurt2D has e Six Wheels e Differential steering e Weight 35 kg In USARSim the Kurt2D is equipped with e One fixed color camera One 2D SICK laser scanner Two encoder sensors One sonar sensor Seven IR sensors The Kurt2D specification is as follows e Dimensions 45cm x 29cm x 35cm e Maximum translate speed 4 m s 12 16 2 Configure it It s the same as the P2AT 12 17 Kurt3D 12 17 1 Introduction KURT3D is a mobile robot of type KURT2D that is equipped with a 3D laser scanner Given the right control and sensor data processing software it is generally able to build 3D models data point clouds of its working environment autonomously In USARSim we use classname USARBot Kurt3D to represent this robot 119 a Real Kurt3D b Simulated Kurt3D Figure 406 Kurt3D Robot In USARSim the Kurt2D is equipped with Two color cameras One 2D SICK laser scanner Two encoder sensors One sonar sensor Seven IR sensors 12 17 2 Configure it It s the same as the P2AT To perform a 3D scan you have to manually set a
14. ia es f i Fa USARSim V3 1 3 A Game based Simulation of mobile robots Prepared by Jijun Wang Edited by Stephen Balakirsky USARSim Contents USARSIM in lis 11 1 Please Tell Us About Your Pro ba is 1 2 44 A dike N N 1 21 BACK round rios 1 222 Whats USARSIM iia 1 A A T E A 3 3 1 SN E 3 3 1 1 Unreal acia 4 3 1 2 Cameo ii 4 LO MOTOR seks io ai cits ROO 4 3 22 Simulator COMPpOnents scsi ee ae ae ee a ae ad awe 5 3 2 1 Environment simulation ssesessseessessseeosserssecrsossenosseressseessoeersosessesesrsse 5 3 22 Sensor and effecter Simulation a cda ini cines 9 3 2 3 Robot SIMU AON Ee aici lh Nas 10 3 2 4 Communications Simul ations josie d 10 3 2 5 Image Server dde 13 3 2 6 MUIIV SW eiii id iii 14 4 Maa a os 20 4 1 A O 20 4 2 Install T2004 iii as 20 4 2 1 WIN WS iia bd ii 20 4 3 MAYS CAN USARSIM oye A A os 22 A A ESTO 23 4 4 Install the con lt i 23 4 4 1 MOAS Vina 24 4 4 2 SAAE OPARE E PEEP EE E E AT E 25 4 4 3 PIYE aie ne mr E earn A 26 5 Run the simulator venida sacd ciedad seta 28 5 1 The steps to run the simulator sssssssesseesseesseeesseeesseesseesseeeseessseeesseesseesseesseee 28 I2 EAS A gh ears tate aaah a N 29 5 2 1 The testing control interface AAA AA 29 5 2 2 MOAS Tis di 30 5 2 3 PTO sc ti 31 5 2 4 Playere a a EA E 32 5 2 5 DIINO UU ae a E A EE see E E A A RS 33 5 3 Getting Starting Poses From Maps sssssessssseesseessessseessseeessetessresseesseeeseeessees 36 5
15. is a number that dictates how many seconds will go by before USARSim drops a Navigation Point Please note that this number is in unreal units and that its default value is 0 Color int string The value for this parameter can be entered as an int or a string It determines the color of the trace as follows 0 or Red Red Trace DEFAULT P or Yellow Yellow Trace 76 2 or Green Green Trace 3 or Cyan Cyan Trace 4 or White White Trace S or Blue Blue Trace 6 or Purple Purple Trace e Query the robot sensor effecter mission package There are two types of query command One queries the geometry information and another queries the configuration information o GETGEO Type string Name string The Name string is optional If it s omitted the command queries the geometry information for all the sensors effecters with the specified type Otherwise only the sensor effecter with the name and type will be queried The return message is a GEO message Example GETGEO Type Sonar GETGEO Type MisPkg GETGEO Type Robot GETGEO Type Effecter o GETCONF Type string Name string Queries the configuration information for a type when Name string is omitted or specified sensor effecter The return message is a CONF message Example GETCONF Type RangeScanner GETCONF Type MisPkg GETCONF Type Robot GETC
16. oooocnoncccnonoccnononcnonenonnnnncnnncnon nnnnnnns 136 1332 Deyice Drivers inan dades 138 Adyanced RA 142 14 1 Build your rin a ise ob 142 E E A EE ERRE 143 141 27 perreo asa E AEEA ee eae San aaa 144 14 1 37 Obstacles and Vic iii iii 144 PAD UU O e e ar be o ORs 146 O O ARN N yea soa esageaa ante so temtaad hme 147 PADD E A E 147 14 2 3 Writing your OWN SEOSOT 5 0sciscsesechevascaytbeccnvestevessancsndecdededunastesaadaceedte 148 14 3 Build your effecter aiii di idad 149 14 3 1 Overview of the Effecter uc Class 4224 acctesiseceascetueecnausaccalsatceteslwescotane 149 14 3 2 Writing your own elec 150 14 4 Build Your robot encinar S E 151 14 4 1 Stepl Build geometric model oooncccnnncccnoncccnonaccnnnacononanononcnonnnncnnnnnnnnnns 151 14 4 2 Step2 Construct the tE a 152 14 4 3 Step3 Customize the robot Optional ocoonnoconnnncccnoncccooncncnonaconnncnnnns 156 14 5 a A alse caaduecgeees oi deceGasvgatnsnesansedunosaae 158 14 5 1 Embedding Unreal Client annonces 158 14 5 2 Capturing Unreal CHEM RR cancers 159 E Sine Ane o AAA O A 160 Information for Gamers acsscicissaasistecivicnss enni eksisi ii aiiai 161 SUS reporto nn sieges cae red ele 161 Conta ii 161 e A a A A A aS 163 vi 1 Please Tell Us About Your Project We are constantly trying to improve USARSim Part of this effort requires us to understand how this package is being used We would greatly appreciate it if you could please fill out a br
17. 14 3 1 Overview of the Effecter uc class In USARSim all effecters are inherited from the Effecter class The Effecter base class defines the valid opcodes and the means to query format information from the various extensions of effecters As mention in Section 8 effecters in USARSim contain a restricted vocabulary of operational codes These opcodes are defined at the top of the Effecter uc file as shown below enum EFFECTOR_OPCODE_TYPE EFFECTOR_OPCODE_ACTIVATE_TYPE EFFECTOR_OPCODE_ANIMATE_TYPE EFFECTOR_OPCODE_FIRE_TYPE EFFECTOR_OPCODE_RELEASE_TYPE EFFECTOR_OPCODE_RESET_TYPE EFFECTOR_OPCODE_NOP_TYPE i The base effecter class implements several virtual functions that will enable the base class to query and send commands to its children There are two basic categories of virtual functions that are implemented in the effecter class the Do lt Opcode Name gt functions and the Get lt Opcode Name gt Conf functions Therefore the developer must implement both functions for each opcode in order to properly implement the opcode These functions will overwrite the virtual functions letting the effecter base class know that these are valid opcodes that can be used for a given effecter The virtual Do lt Opcode Name gt functions defined in Effecter uc These function are used to perform opcodes in the effecter The value specifies the value that is associated with the opcode and the returns a Boolean to indicate whether
18. Apply Texture Reset Alignment 7 A new icon shows up in the world Place it to your desired location and double click on it to bring up the properties Change the ReferenceGPScoordinate properties to desired values Please note that a South direction is achieved by giving negative values to the degree and minutes of the latitude component and that a West direction is achieved by giving negative values to the degree and minutes of the longitude component 85 Advanced Collision Display Events F Movement FE NavigationPoint ES ReferenceGPSCoordinate LatitudeDegree 0 LatitudeMinute 0 000000 LongitudeDegee 0 LongitudeMinute 0 000000 ound e USARBot ini File Open the USARBot ini file and find the USARBot GPSSensor section Under that section add or modify the line ZeroZeroLocation LatitudeDegree 39 LatitudeMinute 8 0273 Lon gitudeDegree 359 LongitudeMinute 59 9996 Please note that a South direction is achieved by giving negative values to the degree and minutes of the latitude component and that a West direction is achieved by giving negative values to the degree and minutes of the longitude component The latitude and longitude components given in that section will provide the GPS coordinate referring to the 0 0 location of the map e GPSSensor uc File If no reference location is found in the map and in the USARBot ini file a default
19. More information on installing cygwin for a MOAST windows installation may be found under the What else do I need topic found at http moast sourceforge net Column 20With 20Contents htm _ What_else_do In addition to the Linux style environment a communications package known as rcs must be installed This is available from the files area of the MOAST sourceforge repository in both source and pre compiled cygwin formats To install the pre compiled version simply download from http www sourceforge net project showfiles php group_id 148555 and then unpack NOTE When unpacking the pre compiled version use the following commands S col S tar BAVIE parm CO resi so coja tes illo 2 Cwenalin o Ez To install the source Install the rcs library at usr local directory cd tar zxvf path_to_rcslib tgz rcslibV0 2 src tgz cd usr local rcslib configure make make install prefix usr local rcslib cd src java make MoM NN YN UY Mm If you want to use the MOAST supplied GUI you need to make sure GTK and GTKI are installed on your machine GTK is a standard package and will be automatically installed during the cygwin installation and is usually installed on Linux operating systems GTKI may be found in the file release area of the MOAST 24 repository It should be downloaded and installed in the usr local sre directory The following commands may be used to install GTKI cd usr local sre ta
20. NIST built three test arenas to help researchers evaluate their robot s performance These arenas are built from the AutoCAD models of the real arenas To achieve high fidelity simulation the textures used in the simulation are taken from the real environment For all of the arenas the simulated environments include Geometric models the model imported from the AutoCAD model of the arenas They are the static geometric objects that are immutable and unmovable such as the floor wall stairs ramp etc Obstacles simulation that simulates the objects that can move and change their states In addition these objects can also impact the state of a robot For example they can change a robot s attitude These objects include bricks pipes rubble etc Light simulation that simulates the light environment in the arena Special effects simulation that simulates the special items such as glass mirrors grid fenders etc Victim simulation is the simulation of victims that can have actions such as waving hands groaning and other distress actions All the virtual arenas are built with Unreal Editor With it users can build their own environment For details please read section 14 1 In addition to the USAR arenas outdoor areas and simulated collapsed buildings have been modeled All of these arenas are available for download at http sourceforge net projects usarsim in the files area The real USAR arenas and simulated arenas
21. Name string string is the sensor name as given in 54 the USARBot ini robot s definition Latitude int float char int float char provide the latitude degree minute as a decimal and cardinal description i e N or S respectively There are only two possible values for the char parameter N for North and S for South Longitude intfloat char int float char provide the longitude degree minute as a decimal and cardinal description i e E or W respectively There are only two possible values for the char parameter E for East and W for West Fix int int indicates whether or not a position was acquired The fix is the same as the GGA format Namely a value of 0 means that the GPS sensor failed to acquire a position and a value of 1 means that a position was acquired Stallites int int gives the number of satellites tracked by the GPS sensor This number is an inexplicit source of accuracy The more satellites are tracked the higher the position accuracy Example SEN Type GPS Name GPS1 Latitude 47 40 3323 N Longitude 122 18 5977 W Fix 1 Satellites 8 o INS Sensor The Inertial Navigation Sensor is a sensor that provides estimates of the vehicles current location and orientation based on measurements of angular velocity and linear acceleration relative to the vehicles current pose SEN Type IN
22. TQ IAtrOdUCIOM aia ico 103 AAA A sacegets sacs sacaeen tease seesaceaaendetes sneeaancecgons 104 IZED ZStereO PZ A NL 106 12221 SMAtROdUCHON Avec A ee Ee E 106 122 2 a AA A 106 123 P2DX uani ae Lads RO A R OR Uisian twain an laltse tat chads Uta dy 106 E SN A O OT 106 123 2 OA A 107 A OEA 107 12 41 gt Introducido ide 107 124 2 Configure itsenne sge n cda 108 12 5 HMMWV Hummetl cccccccccessesessecececececsesenseceeecececeesesssaeceseceesenensaaees 108 UN Introductions sacs etn O E A at 108 123 2 COMMUNE Tdi A S Ei 109 12 6 SNOW OM iii ds o lada 109 IO E AS seie sa os Gis i osse ies Bi os See outa vbotd Ae ish 109 T2625 ME OWING TS 10h 2 5255 Sa cceih cB aenag a vee cadasauaech Seunacnas N a NA 110 DS A O Tacks Bs ARES 110 IET ANtro dictions ia lia sae tae esate 110 iv 127 25 COMM G BTS il a 111 VB ROO DOL eas waren sted wagiseedaay as sigan A O 111 128 1 ntroduU cio aaa 111 A MCA EA UM e e 112 T29 Submarines ri a A RA AAA A AA fot 112 12910 TntrodUc ON soii aliada indias 112 12 9 21 A Gace E ea E seed EEEN vanes EES 113 12 10 Parantulas 2 nne tii 113 12 10 1 Mtro dci o a a re eh ahh E 113 12 10 2 Configuration ao ot 114 12 11 DTG AEEA E EE E EE 114 12 11 1 It OCU CH ON aaa o 114 12 11 2 Configuration iii a et ates E R 115 12 12 A EOI EE E A E E E EEE 115 12 12 1 Introduction id di A ss 115 12 12 2 CONT URC tn E ae ah eae G 115 12 13 ORIO est ho co cts Paves tds
23. The robot model includes chassis parts tires linkage camera frame etc and other auxiliary items such as cameras headlights etc All the chassis and parts are connected through simulated joints that are driven by torques Three kinds of joint control are supported in the robot model The zero order control makes the joint rotate by a specified angle The first order control lets the joint rotate under the specified rotational speed The second order control applies the specified torque on the joint To help better organize and control these parts and joints we introduced a mission package concept A mission package represents a container of parts and joints Sensors and effecters are connected to the robot platform through the mission packages For instance the camera pan tilt frame is a kind of mission package that connects a camera to the robot By controlling the pan tilt frame we can adjust the camera s pose The robot receives the control command and sends out data through Gamebots With this robot model users can build a new robot with little or no Unreal Script programming For the steps of building your own robot please read section 14 3 In USARSim a total of eight ground robots are provided for you P2AT P2DX ATRV Jr Zerg Tarantula Talon Telemax and Soryu In addition USARSim includes four Ackerman steered vehicles for outdoor scenes as well as testing driving algorithms Hummer Sedan SnowStorm and Cooper Two legg
24. Where Type string string is the sensor type It can be either Sonar or IR which means it s a Sonar sensor or IR sensor Name string Range string is the sensor name float is float the range value in meters Example SEN Time 45 14 Type Sonar Name F1 Range 4 4690 Name F2 Range 1 9387 Name F3 Range 1 9159 Name F4 Range 1 6547 Name F5 Range 0 8889 Name F6 Range 0 7640 Name F7 Range 1 1075 Name F8 Range 2 0773 o Laser Sensor SEN Type string Name string Resolution float FOV float Range r1 r2 r3 Where Type string string is the sensor type It can be RangeScanner or IRScanner Name string string is the sensor name Resolution float float is the sensor s resolution in radians With FOV we can calculate the number of the data in the Range segment FOV float float is the sensor s field of view in radians Range rl r2 r3 j rl r2 r3 is a series of range values in meters 53 Example SEN Type RangeScanner Name Scanner Resolution 0 0174 FOV 3 1415 Range 2 2109 2 2075 2 2124 2 2122 2 2156 2 2166 2 2207 2 228 3 2 2334 2 2392 2 2421 2 2533 2 2609 2 2696 2 2806 2 28 94 2 3011 2 3104 2 3243 2 3409 2 3519 2 3708 2 3834 2 4 032 2 4229 2 4429 2 4628 2 4853 2 5051 2 5282 2 5538 2 5837 2 6126 2 6425 2 6696 2 7012 1 3823 1 3642 1 3321 1 3025 1 2733 1 2457 1 2208 1 1951 1 1729 1 1522 1 130 5 1 1109 1 09
25. double arcList 0 speed 0 int arclist 0 isArc 0 4 El gt StatBufferNumber 7 Pict this Command Variable Prior this array Current Status Primo long command_type 7001 202 int echo_serial_number 1 enum RCS_STATUS status RC3_DONE int state 1 int line 263 int source_line 263 char source file 64 src prim primMobMain cc int heartbeat 127813 int pathIndex 0 4 Pict this Status Variable 7 Pior this array Figure 45 MOAST Diag display 7 Use this Status Variable as Figure 45 displays the details page of the diagnostic tool This tool allows you to see the details of the currently executing command for any module and the module s status In addition any command may be sent to any module from this interface This allows for complete unit testing of control code Once again much more information is available on the MOAST website 13 2 Pyro A complete description of Pyro can be found on the Pyro website http pyrorobotics org pyro page PyroModulesContents In this section we only explain the elements that are involved in USARSim 13 2 1 Simulator and world The USARSim simulator loader is put into the pyro plugins simulators directory The loader USARSim py is a Python program that can load the Unreal server and client for the user It reads the world file to figure out which arena map you want Then it will start the Un
26. the parent link number will be 1 Location x y z x y and z gives the x y and z coordinate in meters of the described link relative to its parent s position and based on its parent s orientation Orientation x y z x y and z gives the x y and z orientation in radians of the described link relative to its parent s orientation Please note that the coordinate system is rotated using a ZYX order Example GEO Type MisPkg Name CameraPanTilt Link 1 ParentLink 1 Location 0 1239 0 0000 0 2036 Orientation 3 1415 0 0000 0 0000 Link 2 ParentLink 1 Location 0 0000 0 0000 0 0599 Orientation 1 5707 0 0000 0 0000 e Configuration Information For a ground vehicle the configuration message looks like CONF Type GroundVehicle Name string SteeringType string Mass float MaxSpeed float MaxTorque float MaxFrontSteer float MaxRearSteer float Where Type GroundVehicle Name string SteeringType string Mass float MaxSpeed float MaxTorque float MaxFrontSteer float MaxRearSteer float The Type parameter is hard coded and will always be GroundVehicle for a ground vehicle CONF message string is the robot s name string is one of the following AckermanSteered or SkidSteered or OmniDrive as dictated by the steering type of the robot float is a the robot s mass in kilograms float is the maximum spin speed of t
27. 10 i 4 MaxStram Application Note Indoor Path Loss 91 note the factor 2 means that we are considering forward and backward propagation considering the passive tag as the source of the reflected signal If the received signal is lower than the sensor sensibility dBmRXSensibility then the tag is out of range If there is an obstacle we attenuate the signal by a fixed amount dBObstacleAttenuation In this simplified model we consider only one obstacle This is an acceptable approximation because of the very short range of the passive RFID tags Signal y dBm 40 120 o 2 4 6 12 14 16 18 20 8 10 Distance m Figure 24 Signal strength over distance with dBmTXPower 36 Figure 25 Attenuation mode Traced lines if bTraceRFIDs is true line color represents the signal strength from green to red If the signal is too low the line is black tag is unreachable 10 9 4 Detecting RFID Tags When you are near one or more RFIDTags you will receive the following string SEN Type RFID Name RFID JD 1 Mem 0 if you detect one RFID or SEN Type RFID Name RFID JD 1 Mem 0 ID 2 Mem 0 if you detect more then one at the same time e ID is the unique RFIDTag identifier e Mem Is the RFIDTag memory content 0 if empty This field can be deactivated by setting to false bAlwaysReadRFIDmem 92 10 9 5 Reading RFID Tags Memory To read the memory of the RFIDTag you can use the following command
28. 17 MaxTorque 20 MinRange 0 16 MaxRange 0 40 64 Response Message A response message is delivered to describe the status of a command that has been sent to USARSim Response messages are used so that users can tell whether or not particular commands were successfully executed There are three different response messages The first response message is issued after a SET Type Viewports command The second response messaged is issued after a SET Type Camera command The third response message is a generic message used for sensors and effecters After a SET Type Viewports command the following response message will be issued RES Time float Type Viewports Config string Status string Viewportl string Status string Viewport2 string Status string Viewport3 string Status string Viewport4 string Status string Where Time float float is the timestamp in the virtual world when the message is sent out It is represented in seconds Type Viewports Viewport is hard coded into this parameter Config string string describes the current viewport configuration This parameter will be either Single View or QuadView Status string string is the status of the viewport configuration after the SET command has been issued The status will be OK when the viewport configuration was successfully changed Otherwise the status will be Failed Viewportl string
29. 3 1 A e a a a dang Eaa EOS 36 5 3 2 Adding starting poses to the Map ssssesssessssssesssessssstessresseesseessseesseeesseee 36 5 3 3 Retrieving starting POSES diet ss 43 5 3 4 a A A A cade te aoai 44 5 3 5 Standard Z starting value viii 44 5 3 6 FA E E teasaevraags coasters 45 ii 6 Coordinates Units and Scale ccccccnnnnnonncnnnnnononnnanncnnnncnononananaconoconononenanicicicncnnononos 45 6 1 COOLER A o oe 46 6 2 Units and scale ia 47 A NEO 48 IN A AA as Oe aS OS E TGs SE sn i bate 49 9 Communication amp Control Messages and commands ococcnocccnoncccnoncnonnnnnonancninnnnoss 50 9 1 MAMAS OA 50 92e ThE proto ta 50 9 3 ITS SR ised cede ties a a easy Ges nedeiecay O 51 9 4 Command iaa td obli 67 10 SOS tii ida a E a A a a a 78 TO State Sensorman a a ico S 78 10 1 1 How the sensor works ccccccccccccesssssssecececcceceesenseaeceseccesesenssseaeeeeeceeeeees 78 10 1 2 Howto configure Tus ias 78 10 2 Range Sensor arcada iii 78 10 2 1 How the Sensor works ccccccccccccesessessecececcceceesensaeceeeceesseesesseaeeeeeeeeeenes 78 10 22 TAG WL COMM dure it oopan en Gunegads sovzadhde nte E e 79 10 3 Range Scanner SON id ni ia 80 10 3 L How the sensor works usina rider raider ecco eae 80 103 2 y WPL COMET OUTS MU 65s ees ea E 80 1824 Od metry SENSOR 2c oy aa 81 10 4 1 How the sensor Works ccccccccccccessessssecececcceceesessseceeececseeessnseaeeeeeeeeeees 81 10 4 2 How to configure o da ac
30. 95 10 11 1 How the Sensor WOrkS isura a T A a 95 iii 10 11 2 IOWA confi pure iieiea a E E REES 95 10 12 Human Motion Sensis a e a A a aia aA 96 10 12 1 How the Sensor WoOrKS aoaia a a e a aa 96 10 12 2 Howto CONT BUTS So Ai each Since E EAE 96 10 13 GPS Sensor IA OR Rie aes Pots SORA A 96 10 13 1 How the Sensor WOLKS ccccccccccecesesssssccecesccecsesenseaeceseceeseeessnseaeeesees 96 10 14 Robot Camera ess tt tada eek a leet aes 97 10 14 1 How the Sensor WOLKS ccccccccsccecessesessececesececeesessasececececsceessnseaeeeeees 97 10 14 2 How to COMM SUITE ni eckson each sense ice eco usa R A ves RS 98 10 15 Omnidirectional Camera cccccessessccecccceecessessecececccceessessaeeececeeseeeeseneaaees 98 10 15 1 How the Sensor WOLKS cccccccccccecesssssssecececececeesenseaecesececsesessnseaeeesees 99 10 15 2 Howto TOUTE aiii aiii 99 10 15 3 A O EE 100 11 O EA E ET 100 TLA AAPP dio 100 11 1 1 How the effecter WOTKS ooooocccccnnonnnononnnonocnoncnnonononononcononnnno no noncncononnnos 100 111 2 How t confis re al dlrs deeb tS 101 112 RETIRE learn ad A ISA ag OR Se SES AI 101 Zi How theeffecter Works ainia ios 101 A E ESE 102 11 3 Roller Table o A IE A A BAL OAR 102 11 31 Howtheetfecter Works iii a iia 102 11 3 2 E tah AS 102 MA tiga das baat AE EEE E E EN N E tigate tatunta topes E sateen 103 A A oa toe RO A Re APOC ne Pune SE eee er or OE Rr 103 12 NS 103 A PAT A A A A va 103
31. GGA GPS format modified to keep the structure of the USARSim communication interface Please see the Messages section Section 8 3 of this manual for the specific SEN message employed by the GPS sensor The error model is based on a line of sight signal model that dictates how many satellites can be used by the receiver The number of satellites seen is in turn used to dictate the how much error 82 is injected into the readings The sensor assumes a flat earth where all X axis motion is converted to latitude and all Y axis motion is converted to longitude While in most cases the global X axis points to the North it is worthwhile noting that this is not always the case due to singularities that may occur Indeed the sensor handles singularities that occur at the earth s pole at 90 degree North and 90 degree South and on the longitude at 180 degree West and 180 degree East In other words and as an example driving along the X axis at 89 degrees and 59 9 minutes will increase the latitude component of the GPS until 90 degree North is reached At that point the latitude component will decrease meaning that the global X Axis now points to the south See Figure for an example of the flat world representation at different quadrants of the earth These singularities exist and are taken into account due to the flat assumption of the virtual worlds and the spherical shape of the earth 90N Gicb ATAR z 90S Fig
32. One is using the Unreal Client as a separate sensor panel The other is embedding the Unreal Client into the user interface For details about embedding the Unreal Client into user s application please see section 14 5 1 2 Capturing the scenes in Unreal Client and using these pictures as video feedback This approach is very close to how the real camera works but it s technically difficult The camera feedback can be either directly or remotely received 97 a Directly capture Unreal Client The idea is to capture the pictures in the Unreal Client and use them on the interface There are many capturing technologies We use Detours http www research microsoft com sn detours to access the back buffer of DirectX and get the scene pictures The advantage of this approach is that even if the Unreal Client is hidden is covered by other windows or out of the desktop we can still get the scene image Hook dll available from the tools area of the file release downloads is the library provided by USARSim that captures the Unreal Client picture into a block of shared memory It must be in the UT2004 system directory Details about using Hook dll can be found in section 14 5 2 Using the image server to get camera pictures The Image server is an extra server that runs with the Unreal Client It uses the method introduced in the previous paragraph to capture pictures and send them out through the network The pictures can be sent out
33. The name assigned to this item Position The mounting position relative to its parent Direction The mounting direction relative to its parent uuDirection The reserved variable that stores the direction parameter in unreal units In the robot class Sensors is used for all the sensors Effecters is used for all the effecters And the Cameras stores all the cameras 14 4 3 Step3 Customize the robot Optional After finishing the previous two steps your robot should work You should be able to use the DRIVE command to control every joint and you can also get the sensor data from the robot To go further beyond this you can do three things 1 Write your own control mode 156 The general robot mode only supports controlling every joint separately and two types of mission package control pan tilt and flipper However you can define some control pattern or even control model in your class USARSim uses the DRIVE Left xxx Right xxx command to interact with Pyro The Left and Right means left side and right side wheels separately This is an example of a control pattern In the robot class you can transfer the left right parameters into a series of joint control parameters to control the wheels This can be reached by overriding the ProcessCarInput function of the KRobot class In your own ProcessCarInput you need to call the ProcessCarInput function in KRobot to let your robot interpret the joint
34. USARVictim Properties under the Victim category are the parameters that specify the victim s actions These parameters are AnimTimer Sets how quickly the victim moves Low value means a slow action HelpSound Segments Sets the sound the victim can play Specifies how the body segment moves You can set at most 8 segments For every segment you can define an action The segment will move from the initial pose to the final pose with the specified move rate The action definition parameters are InitRotation FinalRotation PitchRate YawRate RollRate Scale The initial rotation pitch yaw and roll in integer 65535 means 360 degrees of the segment The final rotation pitch yaw and roll in integer 65535 means 360 degrees of the segment The move amount from current pitch angle to the next pitch angle Large PitchRate means tilt quickly It s the same as PitchRate except that it defines the yaw angle It s the same as PitchRate except that it defines the roll angle The scale of this segment 0 will hide this segment Since there is hierarchical relationship in the skeletal system this scale value will affect other segments 145 under it For example hips will affect thigh shine and foot SegName The name of the segment Different skeletal meshes may have different names You can use the Animations browser to view the bone name An example in showed in Figure 51 Intro
35. an example with WideMode 4 19 CameraXres X resolution of robot camera The final resolution of the camera subview will be lt CameraXres If the overall resolution of camera subviews is larger than what can fit in the USARSim client they will be scaled down to fit the screen If for example you use 4 cameras in WideMode 4 each 200 pixel wide and the USARSim client resolution is 640x480 each camera width will be scaled down to 160 pixel But if you set WideMode 0 a different patter is used and the full camera resolution 200 pixel will be used CameraYres Y resolution of robot camera Same as above but for Y axis In any case the aspect ratio CameraXres CameraYres will be preserved That means that a change in resolution will always affect both X and Y IsCameraLocked Don t consider this for now and leave it to false It can be used with zone optimization to speed up subview rendering 3 2 6 4 Limitations There are two important known limitations 1 MultiView supports only one camera can be stereo per robot So if you have a robot that mounts more than one camera like Talon you will see only from its first camera More cameras you use same as saying more robots you use smaller will be the subviews This can or can not be a problem Depends on the specific application 4 Installation 4 1 Requirements Operating System Windows 2000 XP or Linux Software UT2004 with the 3339 or later patch Optional requ
36. and under Linux may all connect simultaneously to these buffers The contents of all of these buffers may be examined by running the RCS Diagnostics tool the third technique This may be run from the moast_base_dir devel src nml directory as either moastDiag csh or moastDiag cygwin csh depending on your operating system 131 RCS Diagnostics 2005 June moast diag servosP servoMis Refresh Time 0 5 seconds Loading sts msgs for primblob l out of l Stop all Plotting Save stace Load State Refresh we w H H Monospaced zo Commands Available PrimMobJACmdInit 7001 PrimMobJACmdHalt 7003 servoMob PrimMobJACmdAbort 7002 Cmd To Send PrimMobJACmdMoveArcSegment 7005 int serial_number 2 double arclist 0 center x 0 PEEOR primsP PrimMobJACmdShutdowm 7004 PrimMobJACmdMoveWaypoint 7006 double arcList 0 center y 0 double arclist 0 center z 0 double arcList 0 end x 0 PrimMobJACmdRotate 7007 PrimMobJACmdV el 7008 PrimMobJACfgCycleTime 7201 double arcList 0 end y 0 double arcList 0 end z 0 double arcList 0 normal x 0 4 PrimMobJAC gMove 7202 PrimMobJACfgDebug 7203 IV connected 2 127813 Current Command PrimMobJACmdInic int serial _number 1 4 double arcList 0 normal y 0 double arcList 0 normal z 0 double arcList 0 theta 0 double arcList 0 annular_tol 0
37. any image processing How to run SimpleUI in local or remote mode is introduced below 5 2 5 1 Using SimpleUI by locally capturing pictures The steps of running SimpleUI are 1 Start Unreal Server see step 1 in section 5 1 2 Execute SimpleUI exe which is located on ut2004 Tools SimpleUI Release directory 3 On the SimpleUI interface set the following parameters and then click the Start button a Command group This group specifies the server that receives the control command Host is the IP address of the Unreal Server Port is the port number of Gamebots By default it s 3000 b Robot group This group defines which robot and where the robot will be added Model is the robot type Position is the location to add the robot Please note in different arenas different position parameters are needed The robots and starting positions are given in a combo box To add new ones make graphical or text changes to SimpleUl rc Text editing this file is obscure since the robot names are not presented as strings c Video group Since we want to get pictures locally we select the Local radio button Resolution sets the picture size Frame Rate sets the maximum video frame rate The actual frame rate is decided by the current computer system s capability 4 The Unreal Client will be launched by SimpleUI and a message box will be popped up to instruct you how to switch t
38. can found all the services The real code for these services is in the robots USARBot __init__ py file The supported sensors are e Sonar io xl Figure 46 Sonar visualization 134 e Laser Figure 473 Laser visualization e PTZ Camera EA ox 90 0 FOW 50 45 0 Zoom In Zoom Qut 0 0 0 5 45 0 Reset 90 0 0 5 300 45 0 00 450 900 Figure 484 PTZ Camera viewer and controller 13 2 4 Brains Pyro refers to control programs as Brains Since the USARSim API follows the Pyro interface the brains of Pyro will work for USAR robots The tested working brains include Slider py Joystick py and BB Wander py 13 3 Player Player provides a client server based hardware abstraction layer Every sensor supports one or multiple sensor interfaces Therefore it s not important if used for example a SICK or a Hokuyo or a simulated laser scanner This provides the possibility to create high level abstract drivers like localization or path planning 135 drivers The following figure shows the UT2004 window and the standard player GUIs playerv and playernav for path planning and sensor data visualisation DX Playerviewer localhost 6665 A de View Dewees Figure 495 playerv sensor data visualization tool top playernav player navigation and path planning tool left the simulated robot in USARSim right In this section we introduce how to control USARSim
39. car wheel joint BrakeTorque The brake torque applied for braking the joint if we are using a car wheel joint Parent The part or chassis the part is connecting to NOTE the part must have already been defined ParentPos The position where the joint connects to the parent ParentAxis For a car wheel joint it s the steering axis relative to the parent For a hinge joint it s the spin axis ParentAxis2 For a car wheel joint it s the spin axis relative to the parent SelfPos The position where the joint connects the part SelfAxis For a car wheel joint it s the steering axis relative to the part For a hinge joint it s the spin axis SelfAxis2 For a car wheel joint it s the spin axis relative to the part The order in which you define the part joint pairs is important Since the parent in the part joint pair must already be defined you need to define the parent before the part You also can define these part joint pairs in the USARBot ini file you may need to create the robot section by yourself By using the USARBot ini file you needn t compile your class after you change something You may find that it s not easy to know the joint position relative to the parent and the part One way to help you figure out these values is using the Unreal Editor At first you put all the chassis and parts in the map in the draw scale you want Then you assemble them together in the map Using some simple geometric objects to represent the
40. current release of the simulation consists of various environmental models levels models of commercial and experimental robots and sensor models As a simulation user you are expected to supply the user interfaces automation and coordination logic you wish to test For debugging and development Unreal spectators can be used to provide egocentric attached to the robot or exocentric third person views of the simulation A test control interface is provided for controlling robots manually Robot control programs can be written using the GameBot interface MOAST System http moast sourceforge net Player interface or Pyro middleware please note that the pyro interface is out of date and not supported 3 System Overview 3 1 System architecture Controller Controller Gamebots Unreal Data 4 gt Control Data Unreal Unreal Engine Models Robots model Sensor model victim model etc Figure 1 System Architecture The system architecture is shown in Figure 1 Below the dashed boxes is the simulator that provides the interactive virtual environment for the users The dashed box is the user side where you can use the simulator to aid in your research The system uses a client server architecture Above the network icon in Figure 1 is the client side It includes the Unreal client and the controller or the user side applications The Unreal client renders the simulated environment In the
41. do is 1 Find a way to place Player Starts in the map all at the same Z value 2 Write a table that says for every robot how much to add or subtract from that Z value Why we need the Player Starts all at the same Z value Because the map the GETSTARTPOSES command doesn t know what kind of robot we are going to spawn therefore it can t adjust the Z value to match our needs The best that it can do 1s to give us some standard Z value We can then use that value to find the correct Z spawning value for our robot The GETSTARTPOSES command could also be modified to take some parameters like the robot type for example but I think it s not a good solution Robot can change over time can be added or deleted we may want to use the same starting positions with different kind of robots etc That s why in my opinion it s better to use some standard Z value It s also possible to modify the INIT command so that it accepts Player Starts tag names as locations But right now I don t have time for that I hope I will return to this solution in the future or maybe someone else can work on it 5 3 5 Standard Z starting value Playing with the editor I discovered that all Player Starts are placed at a standard Z of 47 unreal units that is 18 8cm Also if this value is modified by editing the Player Start position it is reset to 47 after a map re build 44 5 3 6 Z Table Now that we know that the Z value returned by GETSTARTPOSES
42. how to configure the sensor Example CONF Type Camera CameraDefFov 0 8727 CameraMinFov 0 3491 CameraMaxFov 2 0943 CameraFov 0 8726 For an effector the configuration message looks like thisL CONF Type Effecter Name Value Opcode OpcodeName MaxVal Value MinVal Value Where Type Effecter specifies that this message is a configuration message for effecters Name Value Y is the name value pair that indicates the unqiue name of the effecter Opcode OpcodeName is the name of the opcode that the effecter implements Refer to Section 8 for a list of the opcodes MaxVal Value is the name value pair that indicates the upper bound for the value that is accepted by the effecter MinVal Value y is the name value pair that indicates the lower bound for the value that is accepted by the effecter NOTE An effecter can implement more than one opcode and the CONF message returns the configuration information for all effecters on the robotic platform 63 Therefore the segment with the name value pair indicates a new effecter and the opcode name value pair indicates the opcode the effecter specified in by name implements Example CONF Type Effecter Name Roller Opcode Animate MaxVal 6 MinVal 6 For a mission package the configuration information describes each link s properties The message looks like CONF Type MisPkg Name string Link int JointType string
43. in raw format or jpeg format After sending out a picture the server waits for an acknowledgement from the client and then sends out the next picture at the specified frame rate speed The steps to start the image server are listed in section 3 2 5 How to communicate with the server and use the pictures are explained in section 14 5 3 10 14 2How to configure it The robot camera s configuration in the USARBot ini file looks like USARBot RobotCamera Weight 0 4 CameraDefFov 0 7854 CameraMinFov 0 3491 CameraMaxFov 2 0944 Where Weight The weight of the sensor in kg CameraDefFov The camera s default FOV in radians CameraMinFov The minimum FOV in radians CameraMaxFov The maximum FOV in radians If the CameraMaxFov is equal to the CameraMinFov the camera is a fixed FOV camera that can t zoom in or out 10 15 Omnidirectional Camera 98 10 15 1 How the sensor works The omnidirectional Camera is a normal UsarSim camera looking upwards towards a parabolic convex mirror The camera seems to capture the reflection in the mirror depicting the complete 360 surroundings with the perspective distortion that is typical for this type of omnidirectional cameras The creation of a parabolic convex mirror in UsarSim is not possible in a direct way because the underlying Unreal engine only supports planar reflecting surfaces The security camera trick is used to simulate a non planar mirroring surface 0 The camera loo
44. is located The details about how to configure the USARSim devices is explained in section 13 3 2 2 to section 13 3 2 7 The following is an example player configuration file Or position2d 0 QO 0 on driver name us_bot provides simulation 0 port 3000 host 127 0 0 1 pos 5 27 0 5 rot 0 0 0 bot USARBot P2AT botname robotl driver name us_laser provides laser 0 requires simulation laser_name Scannerl driver name us_position provides odometry requires simulation driver name us_sonar provides sonar 0 requires simulation sonar_name F driver name us_sonar provides sonar 1 requires simulation sonar_name R The USARSim robot is defined in simulation 0 It s a P2AT robot The odometry position 0 and laser 0 are USARSim devices The requires 137 simulation 0 specifies that the devices are located on simulation 0 that is the P2AT robot If you want to control multiple robots with player you can insert multiple simulation devices into the configuration file provides simulation 0 provides simulation 1 provides simulation 2 and connect the sensor to a simulation device using requires simulation 0 or requires simulation 1 But it is easier to start a player server for each simulated robot If you want to use the player GUIs pl
45. it is restored under the correct directory If your directory structure looks something like ut2004 ut2004 you need to move all the files under ut2004 ut2004 to ut2004 4 4 Install the controller This step is optional Install a controller only when you want to use MOAST Pyro out of date or Player to control a robot in USARSim 23 4 4 1 MOAST MOAST fully supports USARSim MOAST and all of its related packages may be retrieved from sourceforge Additional packages that must be downloaded include gtki image extensions to gtk used for graphical user interfaces and rcslib an interprocess communications package These packages may be found on the release section of the MOAST site http sourceforge net project showfiles php group_id 148555 The rcslib archive is available as either source or pre compilied for cygwin The MOAST code may be retrieved from CVS To retrieve the code enter the following commands cvs d pserver anonymous cvs sourceforge net cvsroot moast login cvs z3 d pserver anonymous cvs sourceforge net cvsroot moast co P devel Information about accessing this CVS repository may be found in the document titled CVS Version Control for Source Code Alternatively the latest snapshot may be retrieved from the release section of the MOAST site MOAST requires a Linux style environment This may be obtained by running a Linux operating system or by running cygwin under windows
46. joints you can put them on the connection position you want You also may need to assign a name to every object to help you distinguish them After that you can export the map as a t3d file In the t3d file you will find every object s position By subtracting the parent or part s position from the joint position you will get the accurate relative position 155 TIP Assembling the robot in Unreal Editor can help you calculate the relative position Like the real mechanical world improper mechanical structure can cause the robot to be unstable When you create the robot make sure your geometric model is correct You especially need to check if the model has the correct mass distribution In some cases you may need to specify the mass center offset in the KarmaParams When your robot is unstable try to add the parts one by one This can help you figure out which part causes the problem TIP Specifying the mass center offset in the KarmaParams can help you simulate the mass distribution 14 4 2 5 Mount the auxiliary items After you created the robot you can mount other items on it To mount an item please use the following data structure struct sItem var class lt Actor gt ItemClass var name Parent var string ItemName var vector Position var vector Direction var rotator uuDirection where ItemClass The class used to create the item Parent The object on which the item will mount ItemName
47. kenafCameraPan 1 0 kenafCameraTilt 0 0 or MULTIDRIVE kenafCameraPan 1 0 kenafCameraTilt 0 0 Figure 44 Real and simulated Kenaf 12 23 2 Configure it It s the same as the P2AT 128 13 Controller 13 1 MOAST A description of the low level connection the architectural servo and prim levels from MOAST to USARSim is provided here For a more complete description of the MOAST system please refer to the MOAST manual http moast sourceforge net A description of how to install and bring a robot into the environment is presented in Section 5 2 2 It is assumed that you have successfully installed MOAST started the Unreal Server and have run the run script located in the bin directory with SECT VEH and AM set to no and PRIM and USARSIM set to yes NOTE You must set the HOST_NAME in the file moast ini to point to the machine that is running the Unreal Server You should now see the specified robot type at the specified start location in your Unreal Client window if you are running the client The type and location are specified under the block in the moast ini file that pertains to the arena currently under play by the Unreal Server that was connected to Under the MOAST framework all of the sensor data and robot commands are delivered over Neutral Messaging Language NML buffers There are three general techniques for a user or program to interface to these buffers The first is to direc
48. or command is constructed by a data_type and multiple segments data_type and segments are separated by a space Every message or command ends with nn that tells Gamebots the data is finished NOTE Name value pairs are separated by a space Spaces MUST NOT be used elsewhere in the statement For example one must use Location 4 5 1 9 1 8 and not Location 4 5 1 9 1 8 NOTE When you send out a command don t forget to append Inn to tell Gamebots that the data is ended 9 3 Messages There are currently five types of message A State message is the message class that reports the robot or mission package s state Sensor messages contain the sensor data Geometry messages report the robot sensor effecter or mission package s geometry information Configuration messages give the robot sensor effecter or mission package s configuration information and the response message provides a response status to certain command e State and Mission Package message Please note that the robot state message parameters depend on the type of robot that you are driving For example a robot of type GroundVehicle will not have the same state message as a robot of type AerialVehicle A robot state message looks like for a ground vehicle STA Type string Time float FrontSteer float RearSteer float LightToggle bool LightIntensity int Battery int Where Type string string descri
49. rotating the range sensor from the start direction to the end direction in a fixed step The step interval is calculated from the resolution The sensor can work in two modes In the automatic mode it automatically scans data in specified time intervals In manual mode it only works when it gets a scan command and for every scan command it only scans once There are two kinds range scanners RangeScanner and IRScanner The RangeScanner sensor uses the range sensor only emits one detection line to scan the environment While the IR scanner uses the IR sensor the detection line can cross transparent materials to scan the environment Both sensors use the SET command to control the scan behavior We list the Opcode Params and returned response Status below Table 3 Range scanner control command Opcode Params Status SCAN None OK successfully scanned Failed didn t scan It may be caused by an invalid command 10 3 2 How to configure it The RangeScanner and IRScanner sensor share the same configuration format We only list RangeScanner s configuration below For IRScanner the only difference is that the section name should be USARBot IRScanner USARBot RangeScanner HiddenSensor False bWithTimeStamp False Weight 0 4 MaxRange 1000 000000 ScanInterval 0 5 Resolution 800 ScanFov 32768 bPitch false bYaw true Noise 0 0 OutputCurve Points In Val 0 000000 Out V al 0 000000 In V al
50. rotation command to the ScannerSides DRIV of yo E Name ScannerSides Value ROTATION_VALUE If you send this command often enough to GameBots depends on the resolution ur 3D scan you get a 3D scan of the environment This is not as easy as the USARSim RangeScanner3D but using this procedure we can treat the simulated scanner like the real scanner 12 18 Lisa 12 18 1 Introduction The LISA robot is developed in a BMBF funded project during the next 2 years It is an assistance system in a life science environment The physical robot is not built yet We use USARSim to explore the Omni drive behaviour of the robot The robot has two big Omni drive steered wheels on two opposite corners and two small passive whee whee ls on the other corners It is equipped with encoder sensors on the steered ls In USARSim we use classname USARBot Lisa to represent this robot 120 La _ a Real LISA robot b Simulated LISA robot Figure 417 Lisa Robot 12 18 2 Configure it It s the same as the P2AT The LISA script file looks like this Using the Omni drive DRIVE command we are able to drive wheel 0 and in every direction with every speed taking the MaxSteerAngle and MaxSpeed limits into account Wheels 0 Number 0 PowerType Holo_Powered SteerType Holo_Steered MaxSt eerAngle 3 14 Wheels 1 Number 1 PowerType Holo_Powered SteerType Holo_Steered MaxSt eerAngle 3 14 Wheels 2 Number 2 PowerType Not
51. run This file controls which levels of the MOAST hierarchy are automatically started For example setting SECT VEH and AM to no and PRIM to yes will allow control at the level of sending individual wheel velocities Setting VEH to yes will allow waypoints to be sent to the vehicle Full documentation on the levels of control is given on the MOAST webpage For this example we will examine joystick control and waypoint control For joystick control set SECT VEH and AM to no and PRIM to yes Then execute run This will bring up a prim shell that allows for various commands to be sent to the robot and status to be received Typing a carriage return will print the robot status and typing a question mark will show the possible commands at this level Try typing vel 1 0 This will cause the robot to drive in a straight line Another way to control at this level is to run the joystick program To run this open a new window in the bin directory and run joytick Move the mouse into the window that appears and then use the keys r and f to accelerate decelerate the left wheel and the up down arrows to accelerate decelerate the right wheel For waypoint control change the VEH to yes in the run file When this file is executed the PRIM AM and VEH levels will automatically be started The prompt on the screen will be the vehicle shell Once again typing a carriage return will show vehicle status and a question mark will show the possi
52. the University of British Columbia http www ubcthunderbird com for the DARPA Grand Challenge It is a 1991 Jeep Cherokee Laredo modified for autonomous control In USARSim we use classname USARBot Snowstorm to represent this vehicle 109 a Real SnowStorm b Simulated SnowStorm Figure 30 SnowStorm In summary SnowSorm has e Two wheel or four wheel drive Single Ackerman steering front two wheels are Ackerman steered GPS Bumblebee stereo camera A 3D Sick Laser Scanner The SnowStorm specification is as follows e Dimension Length x Width x Height 4 239m x 1 720m x 1 621m 12 6 2 Configure it It s the same as P2AT or Hummer 12 7 Sedan 12 7 1 Introduction The Sedan is an Ackerman steered vehicle based on the Nissan Primera Infinity G20 It was added to USARSim as an additional vehicle for outdoor scenes In USARSim we use classname USARBot Sedan to represent this vehicle 110 a Real Sedan b Simulated Sedan Figure 31 Sedan Vehicle In summary the Sedan has e Four wheels The two front wheels are powered e Single Ackerman steering front two wheels are Ackerman steered The Sedan specification is as follows e Dimension Length x Width x Height 4 2779m x 1 7883m x 1 3238m e Wheel Radius 0 2820 e Maximum translate speed 134 172 kph 83 37mph e Maximum wheel spin speed 100 rad s 12 7 2 Configure it It s the same as P2AT 12 8 Cooper 12 8 1 Introduction The Cooper
53. traces as defined by its configuration section and looks to see if they hit one of the following a victim part e g leg arm head etc or a false alarm e g leg arm head etc The sensor returns all the victim s parts that have been hit including possible false alarms and it is up to the controller user to determine whether the responses indicate a victim or a false alarm Please note that for proper operation the victim sensor should always be mounted on a camera 10 10 2How to configure it The Victim and False Positive Sensor configuration in the USARBot ini file looks like USARBot VictSensor HiddenSensor true Distance 6 HorizontalFO V 0 698 1317 HorizontalStep 0 0698131 VerticalFOV 0 6981317 VerticalStep 0 0698131 bWithTimeStamp true bShowResults false Mean 0 0 Sigma 0 01 Where HiddenSensor Boolean value is used to indicate whether the sensor will be visually shown in the simulator Setting it to true will hide the sensor We recommend setting it to true if it s not necessary to show the sensor When you want to confirm if the sensor is placed in the correct position and has the correct direction you can temporarily set it to false Distance The maximum detection range in meters The traces will be sent Distance meters HorizontalFOV The field of view in the horizontal direction x y plane in radians HorizontalStep The amount of radians between two traces in the horizontal direction x y p
54. turn axis gripper tilt axis and gripper turn axis respectively For example to set the arm in initial pose we use MISPKG Name TeleMaxArm Link 1 Value 0 Link 2 Value 2 Link 3 Value 0 Link 4 Value 2 Link 5 Value 0 Link 6 Value 0 Link 7 Value 0 Flipper control MULTIDRIVE FRFlipper x FLFlipper y RRFlipper z RLFlipper r where x y z r are the flipper angles in radians The four parameters control the front left front right rear left and rear right track separately For example to stow the tracks to lift or lower the chassis we use MULTIDRIVE FRFlipper 1 5 FLFlipper 1 5 RRFlipper 1 5 RLFlipper 1 5 or MULTIDRIVE FRFlipper 1 5 FLFlipper 1 5 RRFlipper 1 5 RLFlipper 1 5 Gripper control SET Type Gripper Name Gripper Opcode Open Params x where x is the desired open angle in radians For example to open and close the gripper we use SET Type Gripper Name Gripper Opcode Open Params 1 and SET Type Gripper Name Gripper Opcode Open Params 0 12 19 2 Configure it It s the same as P2AT 12 20 AirRobot 12 20 1 Introduction The AirRobot is a four rotor electrical helicopter with flight stabilization control produced by AirRobot Co http www airrobot com englisch index php The AirRobot serves many purposes including but not limited to exploration observation documentation and measurement In USARSim we use classname USARBot AirRobo
55. value is a tick count which is the real angle divided by the sensor s resolution How we mount the sensor will decide what axis s spin angle will be measured The sign of the count is also decided by the direction we mount the sensor For example mounting the sensor on the front or back side of a wheel will give us a different count sign Similar to the INU sensor we can use the SET command to reset the tick count The Opcode Params and returned response Status are listed below Table 5 Encoder sensor control command Opcode Params Status RESET None OK successfully reset Failed didn t reset the sensor It may be caused by an invalid command 10 7 2 How to configure it We configure the sensor in the USARBot ini file The configuration looks like USARBot EncoderSensor HiddenSensor true bWithTimeStamp False Weight 0 4 Resolution 0 01745 Noise 0 005 Where HiddenSensor Boolean value is used to indicate whether the sensor will be visually shown in the simulator Setting it to true will hide the sensor We recommend setting it to true if it s not necessary to show the sensor When you want to 88 confirm if the sensor is placed in the correct position and has the correct direction you can temporarily set it to false bWithTimeStamp Indicates whether the timestamp in the Unreal Engine will be associated with the sensor data Weight The weight of the sensor in kg Resolution The min
56. 0 220 80 150 240 25 170 240 25 170 y 220 80 150 185 115 130 145 130 90 12 1 2 Configure it The whole P2AT robot configuration can be found in the section USARBot P2AT of USARBot ini file The following lists the parameters you may want to change For other parameters please refer section 14 4 2 USARBot P2AT msg Timer 0 200000 bAbsoluteCamera true bMountByUU True 104 Weight 14 Payload 40 batteryLife 3600 Sensors ItemClass class US ARBot SonarSensor ItemName F1 Position X 7 6125 Y 6 825 Z 0 Direction Y 0 Z 16384 X 0 Sensors ItemClass class USARBot SonarSensor ItemName F8 Position X 7 6125 Y 6 825 Z 0 Direction Y 0 Z 16384 X 0 Cameras ItemClass class USARBot RobotCamera ItemName Camera Pare nt CameraTilt Position Y 0 X 4 2 Z 3 36 Direction Y 0 Z 0 X 0 Where msgTimer The time interval between sending two consecutive messages bAbsoluteCamera Indicates whether the camera control command uses an absolute value or not bMountByUU Indicates whether we use unreal units in mounting sensors If it s true all the sensor effecter position and direction parameters are in Unreal Unit Weight The weight of the chassis in kg Similar to sensor s weight it s just an attribute for description purposes Payload The robot s payload capability in kg batteryLife The life of the battery in seconds Sensors The sensors mounted on the robot T
57. 0 will drive the robot backward with a spin speed of 1 rad sec DRIVE Speed 1 0 FrontSteer 0 523599 will drive the robot 30 forward and to the left with a spin speed of 1 rad sec DRIVE Speed 1 0 FrontSteer 0 523599 will drive the robot 30 forward and to the right with a spin speed of 1 rad sec o DRIVE Propeller float Rudder float SternPlane float Normalized bool Light bool Where Propeller float is the spin speed for the propellers If we use float normalized values the value range is 100 to 100 and corresponds to the propeller s minimum and maximum spin speed respectively Otherwise the value is the absolute propeller s spin speed in radians per second Rudder float specifies the angle of the robot s rudders If we float use normalized values the value range is 100 to 100 and corresponds to the rudders minimum and maximum steer angle respectively Otherwise the value is the absolute rudder angle in radians SternPlane float specifies the angle of the robot s stern planes If float we use normalized values the value range is 100 to 100 and corresponds to the stern planes minimum and maximum angle respectively Otherwise the value is the absolute stern plane angle in radians Normalized Indicates whether we are using normalized values or bool not The default value is False which means absolute values are used Light bool bool is whether turn on o
58. 004_Build_ 2005 11 23_ 16 22 F Games UT2004 PIE File Edit View Brush Build Tools Help PlayerStart E Any map must have at least 1 PlayerStart to be a valid map You can add as many PlayerStarts as you want Follow these steps to add one 37 tf Unreal Level Editor Build UT2004_Build_ 2005 11 23 _ 16 22 F Games UT2004 Mi E3 File Edit View Brush Build Tools Help sa DLs 15 IV Use Actor as Parent IV Placeable classes Only E Actor amp Cinematic_Camera E AmbientSource AntiPortalctor Decoration Emitter FluidSurfaceO scillator Info lonCannon E KActor ORV Rl AabarD anbar In the tree view find Actor NavigationPoint SmallNavigationPoint PlayerStart and select it 38 Unreal Level Editor Build UT2004_Build_ 2005 11 23_ 16 22 F Games UT2004 FRIES File Edit View Brush Build Tools Help File View Class Sa 0 4 1 IV Use Actor as Parent IV Placeable classes Only LiftCenter LiftE xit PathNode ShootSpot E SmallNavigationPoint Ladder E i Teleporter Network Trigger Note 0ONSTeleportPad fon ME ee Now right click in the 3D map and chose Add PlayerStart Here 39 tf Unreal Level Editor Build UT2004_Build_ 2005 11 23 16 22 F Games UT2004 Mi E3 File Edit View Brush Build Tools Help f Actor Classes Engine PlayersS Mila Eg File View Class Texture
59. 1000 0000 00 OutVal 1000 000000 Where HiddenSensor Boolean value is used to indicate whether the sensor will be visually shown in the simulator Setting it to true will hide the sensor We recommend setting it to true if it s not necessary to show the sensor When you want to confirm if 80 the sensor is placed in the correct position and has the correct direction you can temporarily set it to false bWithTimeStamp Indicates whether the timestamp in the Unreal Engine will be associated with the sensor data Weight The weight of the sensor in kg MaxRange The maximum distance that can be detected ScanInterval It is the time interval between scanning used in automatic mode Resolution The scan resolution the step length of rotating from start direction to the end direction The unit is integer 65535 means 360 degrees ScanFov The scan field of view as an integer 65535 means 360 degrees bPitch Boolean value that indicates the scan plane True means scanning in the tilt plane x z plane bYaw Boolean value that indicates the scan plane True means scanning in the pan plane x y plane Noise The relative random noise amplitude With the noise the data will be data data random noise data where random noise returns a value between noise and noise OutputCurve The distortion curve It is constructed by a series of points that describe the curve Note Too much sensor data will impact the system Do not set the res
60. 175 45 172 49 178 6 60 117 2 181 1 75 72 26 181 1 90 295 17 181 1 90 guessed data 347 06 150 36 180 guessed data 347 06 150 36 180 guessed data 295 17 181 1 90 guessed data 72 26 181 1 90 117 2 181 1 75 172 49 178 6 60 230 23 175 45 12 4 2 Configure it It s the same as P2AT 12 5 HMMWV Hummer 12 5 1 Introduction The HMMWV is a High Mobility Multipurpose Wheeled Vehicle built by the National Institute of Standards and Technology The HMMWV is a test bed vehicle used to test evaluate and demonstrate advanced mobility technology at test facilities and on real roads while performing transportation specific operations In USARSim we use classname USAR Bot Hummer to represent the HMMWV 108 a Real HMMWV b Simulated HMMWV Figure 29 HMMWV Vehicle In summary the HMMWV has e Four drive wheels e Single Ackerman steering front two wheels are Ackerman steered In our simulation it is equipped with e A pan tilt camera that can take a 360 degree panorama e A Sick Laser Scanner LMS200 The HMMWY specification is as follows e Dimension Length x Width x Height 3 686m x 1 799m x 2 059m e Wheel Radius 0 3727m e Maximum translate speed 134 172 kph 83 37mp e Maximum wheel spin speed 100 rad s 12 5 2 Configure it It s the same as P2AT 12 6 SnowStorm 12 6 1 Introduction SnowStorm is the vehicle used by
61. 22 1 0741 1 0568 1 0412 1 0266 1 0110 0 99 76 0 9845 0 9714 0 9592 0 9474 0 9377 0 9269 0 9170 0 9 074 0 9001 0 8906 0 8835 0 8757 0 8680 0 8622 0 8556 0 8483 0 8440 0 8376 0 8331 0 8288 0 8248 0 8206 0 8070 0 8155 0 9298 1 0322 1 1549 1 3122 1 5795 1 7881 1 795 2 1 7916 1 7913 1 7909 1 7891 1 7879 1 7878 1 7880 1 78 98 1 7923 1 7960 1 8005 1 8023 1 8063 1 8131 1 8189 1 8 236 1 8328 1 8395 1 8468 1 8541 1 8658 1 8740 1 8874 1 8990 1 9084 1 9234 1 9373 1 9507 1 9686 1 9843 1 9991 2 0188 2 0387 2 0568 2 0787 2 1020 2 1261 2 1519 2 178 2 2 2026 2 2309 2 2636 2 1613 2 2169 2 2754 2 3380 2 40 81 2 4626 2 5146 2 5325 2 5659 2 5955 2 6276 2 6645 1 5 562 1 5928 1 6312 1 6698 1 7121 1 7554 1 8054 1 8507 1 9076 1 9653 2 0274 2 0935 2 1908 2 2722 2 3553 2 3809 4 5273 4 7421 4 9283 4 7977 4 6330 4 4815 4 3422 4 211 4 4 0851 3 9718 3 8664 4 7144 4 6967 4 6828 4 6715 4 65 71 4 6439 4 6354 4 6221 4 6157 4 6093 4 6163 4 6132 4 6 037 4 5944 o Odometry sensor SEN Type Odometry Name string Pose x y theta Where Name string string is the sensor name Pose x y theta x y is the estimated robot position relative to the start point in meters theta is the head angle in radians relative to the start orientation Example SEN Type Odometry Name Odometry Pose 0 2415 0 0029 0 5157 o GPS Sensor SEN Type GPS Name string Latitude int float char Longitude int float char Fix int Satellites int Where
62. 5 In the tree view search for MultiView and select it The path to it is Actor Pawn MultiView Now right click somewhere in the map in the perspective view and select Add MultiView Here This is shown in Figure 11 Unreal Level Editor Build UT2004_Build_ 2005 11 23_ 16 22 F Games UT2004 Maps DM Mi Eg File Edit View Brush Build Tools Help alsa E Actor Classes USARBot MultiView Mi E3 File View Class Textures Actor Classes Meshes Animations Sta4 gt 3 ola Ol le Edit Add Mu ew Here Add Static Mesh 2k4 ae rgerMeshes ChargerMeshes HealthChargerMESH DS Add Path Node Here Add Player Start Here Add Light here Grid Pivot Level Properties USARRobot UnrealPawn Vehicle worldController Pickup ProjectileS pawner Figure 11 Adding the multiview pawn Close the Actor Classes view The box shown in should have appeared in the map f Unreal Level Editor Build UT2004_Build_ 2005 11 23_16 22 F Games UT2004 Maps DM olx File Edit View Brush Build Tools Help Figure 12 MultiView object added to map 16 You can place it wherever you want for example near the ceiling so to hide it from robot cameras Now click on the rebuild button to rebuild the map as shown in Figure 13 and save the map tf Unreal Level Editor Build UT2004_Build_ 2005 11 23_ 16 22 F Games UT2004 Maps DM B ES File Edit View Brush Build Tool
63. A xvoting u E urzK4 u uT2k4Assaut u Axweapons u R 1pDrw u JuTzk4AssaultFull u E XWeapons_rc u E onslaught u Eurclassic u EJ xwebadmin u E OnslaughtsP u Jurv2004c u R OnslaughtFul u uTv2004s u SkaarjPack u Juweb u My Network File name USARBotu Places Files of type lass Packages u Cancel 5 Navigate to Actor gt NavigationPoint gt PathNode and Select ReferenceGPSCoordinate My Documents My Computer 84 Actor Classes USARBot ReferenceGPSCoordinate DER Ele view Class Textures Actor Classes Meshes Animations Static Meshes Prefabs Groups Sounds Music a sil Oe IV Use Actor as Parent IV Placeable classes Only E Actor ES Cinematic_Camera E AmbientSource AntiPortalActor se Decoration e Emitter FPSLog FluidSurfaceOscilator Info lonCannon KActor KNActorFactory Mask BlockedPath Door JumpDest JumpPad KarmaBoostDest LiftCenter Tracing ShootSpot SmallNavigationPoint NetworkTrigger 6 Right Click the location that you want to give a GPS coordinate in the 3D view of the UnrealEd and select Add ReferenceGPSCoordinate Here Surface Properties 1 Selected At SertaMestrlahehaheryerMeetes ChargerMeshes HealthChargerMESH DS Add Path Node Here Add Player Start Here Add Light Here Select Surfaces Select All Surfaces Shift S Select None Shift Z Extrude Bevel
64. ARPA urban challenge robotic soccer submarines humanoids and helicopters USARSim is designed as a simulation companion to the National Institute of Standards NIST Reference Test Facility for Autonomous Mobile Robots for Urban Search and Rescue Jacoff et al 2001 The NIST USAR Test Facility is a standardized disaster environment consisting of three scenarios Yellow Orange and Red physical arenas of progressing difficulty The USAR task focuses on robot behaviors and physical interaction with standardized but disorderly rubble filled environments USARSim supports HRI by accurately rendering user interface elements particularly camera video accurately representing robot automation and behavior and accurately representing the remote environment that links the operator s awareness with the robot s behaviors High fidelity at low cost is made possible by building the simulation on top of a game engine By offloading the most difficult aspects of simulation to a high volume commercial platform which provides superior visual rendering and physical modeling our full effort can be devoted to the robotics specific tasks of modeling platforms control systems sensors interface tools and environments These tasks are in turn accelerated by the advanced editing and development tools integrated with the game engine leading to a virtuous spiral in which a widening range of platforms can be modeled with greater fidelity in less time The
65. ARSim e Proprioceptive sensors These include battery state and headlight state e Position estimation sensors These include location rotation and velocity sensors e Perception sensors These include sonar laser pan tilt zoom ptz camera touch sensor and RFID tag reader All of the sensors in USARSim are configurable A sensor can be easily mounted on the robot by adding a line into the robot s configuration file When a sensor is mounted its name type position where it s mounted and the direction it will face can be specified For every kind of sensor specific properties can be specified Examples of these include the maximum range of the sonar the resolution of the laser and FOV field of view of the camera For more information about configuring a sensor please see section 10 For details of mounting a sensor on the robot please see section 12 Effecters are very similar to sensors They can be configured and mounted on the robot However instead of sending sensor data to the user the main function of an effecter is to accept a command and execute the corresponding function in the virtual world Currently only headlights and RFIDReleaser effecters exist The details of effecter can be found in section 0 How to equip an effecter is explained in section 12 3 2 3 Robot simulation Using the Karma rigid body physics engine which is embedded in Unreal Tournament 2004 we built a robot model to simulate the mechanical robot
66. GPS coordinate will be used for the 0 0 map location That GPS coordinate is defined in the defaultproperties section of the GPSSensor uc file 10 6 INS Sensor How the sensor works The INS Sensor implemented in USARSim uses the Gaussian random number generator found in USARUtils uc to add noise to the ground truth reading of angular velocities and distance traveled per time step These values are then used to estimate the robot s current location and orientation Depending on the configuration errors may cumulate and therefore diverge slowly from the ground truth General description of how algorithm works 1 Calculates angular velocity from ground truth 2 Uses Gaussian random number generator to add noise to each of the angular velocities components 3 Uses these angular velocity components to update the vehicle s estimation of the current orientation 4 Calculates the total distance traveled in one time step 5 Adds Gaussian noise to distance 6 Decomposes the distance traveled in that time step into distance vectors using polar coordinates transforms on the current estimation of the vehicle s orientation 7 Add these estimates to the estimates of the vehicle s location 86 NOTE All noise is proportional to rate of change more change causes more error The INS sensor uses USARSim ground truth for location and orientation as the initial estimate of the robot s location and orientation This sensor contains a SET
67. IVE XZAngle 0 5 will rotate the support motor bars 0 5 radians DRIVE ThrustPropeller 1 will thrust the robot at a rate of 1 meters per second DRIVE TailPropeller 0 3 will rotate the robot to the right at a rate of 0 3 radians per second 12 22 Rugbot The Rugbot Robot is a compact lightweight rugged vehicle developed at Jacobs University Bremen It is ideal for both autonomous and teleoperated mode of operation In USARSim we use classname USARBot Rugbot to represent this robot a Real Rugbot b Simulated Rugbot Figure Rugbot Robot In summary a Rugbot has e Two tracks e Weight 34 kg 126 e Payload 45 kg In USARSim the Rugbot is equipped with e One Panasonic Pan Tilt camera e One SICK LMS sensor e One odometry sensor e One INU sensor The Rugbot specification is as follows e Dimension Length x Width x Height 38 5cm x 25 5cm x 27 0cm e Wheel radius 0 125cm e Maximum wheel spin speed 6 0 rad sec Configure it It s the same as P2AT 12 23 Kenaf 12 23 1 Introduction The Kenaf is a 6 track mobile robot platform which is designed for drastic performance gain of uneven terrain mobility In fact the Kenaf wins championship in RoboCup Rescue 2007 Mobility Challenge Each track of the robot grasps the environment firmly and makes the robot to be grounded to the environment and the robot gets over the unknown rough terrains This robot will be commercially available In USARSim we use classname
68. ONF Type Effecter e Manage viewpoint SET Type Camera Robot string Name string Client ip This command sets the viewpoint of the specified Unreal Client to a robot s camera The unreal client is defined by Client ip where ip is the client s IP address Please note USARSim doesn t support the loopback ip address So don t use 127 0 0 1 as the parameter The robot is specified by Robot string where string is the robot s name And the camera is specified by Name string where string is the camera s name Once the client s viewpoint is set we can NOT manually change it until we release the viewpoint control To release the control we send another SET command without robot name For example we can send SET Type Camera Client 10 0 0 2 to release the viewpoint control on client 10 0 0 2 We can use this command at anytime and anyplace This command can be sent either from a robot s controller or from other applications such as the ImageServer NOTE USARSim doesn t support the loopback ip address Please don t use 127 0 0 1 as the parameter 17 10 Sensors In USARSim every sensor is an instance of a sensor class a sensor type All of the objects of a sensor class have the same sensor capability You can configure the capability of a sensor class to satisfy your needs or you can create a new sensor from an existing sensor class and change
69. Pan and Tilt Head Camera 12 13 2 Configure it It s the same as the P2AT 12 14 ERS 12 14 1 Introduction The ERS also known as AIBO is an artificially intelligent robotic pet designed and manufactured by Sony The ERS is able to walk see its environment via camera and recognize spoken commands ERS robots are used in the Four Legged League of RoboCup soccer In USARSim we use classname USARBot ERS to represent this robot 116 a Real ERS b Simulated ERS Figure 37 ERS Robot In summary the ERS has Four legs each of which has three joints Paw sensors in each Leg Head made of three joints Head Camera IR Distance Sensors Near and Far Weight 1 6 kg The ERS specification is as follows e Dimensions 31 9 cm x 18 0 cm x 27 8 cm 12 14 2 Configure it It s the same as the P2AT 12 15 Soryu 12 15 1 Introduction The IRS Soryu is a serpentine robot used for search and rescue developed by Tokyo Institute of technology and International Rescue System Institute It has the ability to easily intrude narrow spaces as well as traverse rough terrain In USARSim we use classname USARBot Soryu to represent this vehicle 117 a Real Soryu b Simulated Soryu Figure 38 Soryu Robot In summary the Soryu has e Three cars each of which has two tracks Two joints between each car pan and tilt CCD and IR cameras Ability to recover from lying on its side Weight 10 kg The ERS specification is as follows
70. S Name string Location x y z Orientation r p y Awe Where Name string string is the sensor name Location x y z x y z are float variables for estimated vehicle locations 55 Orientation r p y r p y are float variables for estimated vehicle orientation in radians All radians are in the range 0 2PI Example SEN Type INS Name INS Location 1 23 2 23 0 12 Orientation 4 57 3 14 1 507 Example SEN Type Encoder Name ECLeft Tick 61 Name ECRight Tick 282 Name ECTilt Tick 0 Name ECPan Tick 0 o Encoder sensor SEN Type Encoder Name string Tick int Name string Tick int Where Name string Tick int string is the sensor name int is the tick count Example SEN Type Encoder Name ECLeft Tick 61 Name ECRight Tick 282 Name ECTilt Tick 0 Name ECPan Tick 0 o Touch sensor SEN Type Touch Name string Touch bool Name string Touch bool Where Name string Touch string is the sensor name bool bool indicates whether the sensor is touching something Value True means the sensor is touching something Example SEN Type Touch Name Front Touch True Name Left Touch False Name Right Touch False o RFID sensor RFID tags are simulated by implementing the class USARBot RFIDTag class in the Unreal Editor when editing a map This class contains an integer id and a boolean bSingleshot variable that determines wh
71. T_NAME located under the USARSIM_API section This should be set to the host that is running the unreal server When the system is started it will communicate with the Unreal Server to determine which world is in play and will then read that world s parameters These parameters the second interesting item are located in the section WorldName where worldName is the name of the world in play and have the following meaning UNREAL_UTM_OFFSET This provides an offset between the location reported by the simulation and the location reported by the robot over its navigation channel It is used to georeference arenas to the real world and utilizes the Universal Transverse Mercator UTM coordinate system The offset is provide as a triplet of northing easting down UTM_LETTER MOAST utilizes the letter S for the southern hemisphere and N for the northern hemisphere UTM_ZONE The UTM zone The zone locations may be found at http www dmap co uk utmworld htm UTM_START_POSE_COUNT The number of start poses that are included in this section UTM_START_POSE_x The starting location of the robot in offset coordinates 30 NOTE If the world being used is fictitious then the UNREAL_UTM_OFFSET should be set to 0 0 0 and the letter and zone may be set to any value 5 2 2 2 Running MOAST The run script for starting MOAST is located in the dev bin directory under the MOAST home directory The file is named
72. Tools Help ai n DE ME This way you can chose standard IDs for every starting pose that will allow you to chose the right one Now save the map and close the editor It s not necessary to compile the map because you don t need to create a weighted bot navigation graph 5 3 3 Retrieving starting poses Now when you connect to USARSim you can send this command before spawning any robot GETSTARTPOSES You will receive a string like this one NFO StartPoses 3 BlueGoal 2 55 0 00 0 19 0 00 0 03 0 01 MiddleField 0 01 0 00 0 19 0 00 0 00 0 00 YellowGoal 2 57 0 01 0 19 0 00 0 00 3 13 The syntax of this string doesn t follow strictly the USARSim standard I decided for it only because it s simpler to parse If you don t like it it can be changed easily 43 StartPoses is the number of starting positions available For every starting position you ll receive it s tag location and orientation BlueGoal 2 55 0 00 0 19 0 00 0 03 0 01 where BlueGoal is the tag 2 55 0 00 0 19 is the location and 0 00 0 03 0 01 is the orientation If there are no starting positions defined that s bad because a map must have at least 1 PlayerStart you ll receive NFO StartPoses 0 5 3 4 Start Elevation As we want to improve spawning precision we don t want robots to jump from the ground or fall from the sky when created we need to have exact Z spawning values for the robots we are using What we have to
73. Unreal client through changing the viewpoint we can get the view of the robot in the environment All the clients exchange data with the server through the network The server side is called the Unreal server It includes the Unreal engine Gamebots the map and the models such as robot models victim models etc The Unreal server maintains the states of all the objects in the simulator responds to the data from the clients by changing the objects states and sends back data to both Unreal clients and the user side controllers In summary the three main components that construct the system are 1 the Unreal engine that makes the role of server 2 the Gamebots that communicates between the server and the client and 3 the Control client that controls the robots on the simulator 3 1 1 Unreal engine The Unreal engine used in the simulator is released by Epic Games http www epicgames com with Unreal Tournament 2004 Please note that a full license for the Unreal Tournament 2004 game is required cost of about 40 at most software retailers http www unrealtournament com ut2004 The demonstration version will not work with USARSim It s a multiplayer combat oriented first person shooter for the Windows Linux and Macintosh platforms In addition to the amazing 3D graphics provided by the engine the physics engine which is known as the Karma engine is also included in Unreal to obtain high quality reality Unreal engine also p
74. VE Name string Steer int Order int Value float Where Name string string is the joint name Steer int inf is the steer angle of the joint Order int int is the control mode It can be 0 2 0 zero order control It controls rotation angle 1 first order control It controls spin speed 2 second order control It controls torque Value float float is the control value For zero order control it s the rotation angle in radians For first order control it s the spin speed in radians second For second order control it s the torque Example DRIVE Name LeftFWheel Steer 1 57 will steer the left front wheel 90 degrees DRIVE Name LeftFWheel Order 1 Value 0 175 will make the left front wheel spin at 0 175 radians second 1 e 10 degrees second o MULTIDRIVE string float string float Where string float The string float pair describes a joint to move and a position to move to string is the name of a joint as described in the USARBot ini file float is the absolute angle in radians that the joint should move to Please note that any number of string float pairs can be sent using a single MULTIDRIVE command to move multiple joints at the same time Example MULTIDRIVE FRFlipper 1 FLFLipper 1 will move the two front flippers to 1 radians MULTIDRIVE RRFlipper 1 RLFLipper 1 will move the two rear flippers to 1 radians e Control a j
75. We recommend setting it to true if it s not necessary to show the sensor When you want to confirm if the sensor is placed in the correct position and has the correct direction you can temporarily set it to false Weight The weight of the sensor in kg MaxRange The maximum detecting range in meters FOV The field of view of the sensor in integer 65535 means 360 degrees Noise The relative random noise amplitude With the noise the data will be data data random noise data where random noise returns a value between noise and noise OutputCurve The distortion curve It is constructed by a series of points that describe the curve 10 13 GPS Sensor 10 13 1How the sensor works The GPS sensor finds the position in meters and then converts it to latitude and longitude It assumes a flat earth and assumes that the Y axis points due north All Y axis motion is converted to latitude All X axis motion is longitude The 96 sensor does not check for an open sky and may thus report GPS when it would not be available Any robot can navigate from relative GPS Relating the absolute GPS to the map requires an RNDF For RNDF specifications see http www darpa mil grandchallenge docs RNDF_MDF_Formats_120606 pdf DARPA s Dec 1 2006 revision specifies that latitude and longitude are fixed point numbers in the ITRFOO reference frame The current RNDF uses the previous specification which gives latitude and longitude as floating point
76. _Powered Wheels 3 Number 3 PowerType Not_Powered 12 19 TeleMax 12 19 1 Introduction The TeleMax is an EOD explosive ordnance disposal robot built by Rheinmetall Defence http www rheinmetall defence com that intends to work in narrow spaces In USARSim we use classname USARBot TeleMax to represent this robot 121 a Real TeleMax b Simulated Telemax Figure 428 Telemax Robot In summary a TeleMax has e Four tracks e One 7 axis manipulator with turret and linear axis The TeleMax specification is as follows Wheel Diameter x Width 38 1 cm x 9 5 cm Size Length x Width x Height 80 cm x 40 cm x 75 cm Stowed Position Vertical reach 235 cm Horizontal reach front 120 cm Maximum Speed 1 3 m s Operation Time 2 hours In our simulation it is equipped with e Fore fixed color camera e One gripper with two fingers e One Odometry sensor e One INU sensor To control the arm in USARSim we use the following commands Arm control MISPKG Name TeleMaxArm Link 1 Value x Link 2 Value y Link 3 Value z Link 4 Value r Link s Value t Link 6 Value u Link 7 Value v 122 where x y z r S t u v are either the rotation angles in radians for revolute joints or translation distance in meters for prismatic joints From the turret to the arm terminal the gripper Link 1 7 correspond to the turret pan axis upper arm tilt axis telescope lower arm tilt axis lower arm
77. accesses the laser whose name is the same as the name specified in the configuration file Interfaces Supported interfaces e laser Required devices e simulation Configuration file options Name Type Default Meaning simulation int 1 The simulation id that specifies the USARSim robot where this device is located Laser_name string The name of the laser 13 3 2 6 us_fakelocalize Synopsis The us_fakelocalize driver is used to provide a ground truth localization in player using the USARSim STA message Interfaces Supported interfaces e position2d 140 Required devices e simulation Configuration file options Name Type Default Meaning simulation int 1 The simulation id that specifies the USARSim robot where this device is located origin int int int 0 0 0 This value is added to the USARSim position to adjust the USARSim coordinate system to the player coordinate system Only x and y are used 13 3 2 7 us_ptz Synopsis The us_ptz driver is used to control the robot s ptz camera But you can t get the current orientation and camera velocity yet The camera should use absolute pose control Interfaces Supported interfaces e ptz Required devices e simulation Configuration file options Name Type Default Meaning Simulation int Ki The simulation id that specifies the USARS
78. and could not be carried out When GETSS is successfully carried out the response is OK PathLoss Where PathLoss is the path loss between the robots in dBm If it fails the response is a fail message as described above 3 2 4 4 Sample Programs Very simple sample programs using the server can be found in the USARSim CVS repository http sourceforge net cvs group_id 145394 under the folder Tools WSS fic Image Server UT Startup UT Client Mode UT Server address IP UT Server Mode An UT Map Image Settings Connected Clients Resolution 320x240 Clients 160x120 480x360 a Frame Rate 110 Figure 8 Image server control interface 3 2 5 Image Server The image server is a wndows application that allows one to capture images from any of the cameras that are being used inside of the game engine It may be run with UT running in either client or server mode The control display from the image server may be seen in Figure 8 13 3 2 5 1 Installation The image server may be obtained from CVS or from the downloads page under the Tools area on Sourceforge as an install package 3 2 5 2 Running the Image Server with USARSim in Server Mode If you want to use the image server in the usual way do the following 1 Start the UT server Start the image server 3 In the image server select UT Server Mode type in UT Server address IP enter a map name and chose the format resolution and
79. are listed below The yellow arena the simplest of the arenas It is composed of a large flat floor with perpendicular walls and moderately difficult obstacles Figure 2 Yellow arena Figure 3 Simulated yellow arena The orange arena a bi level arena with more challenging physical obstacles such as stairs and a ramp The floor is covered with debris including paper pipes and cinder blocks Figure 4 Orange arena Figure 5 Simulated orange arena The red arena presents fewer perceptual difficulties but places maximal demand on locomotion There are rubble piles cement blocks slabs and debris on the floor Figure 7 Simulated red arena 3 2 1 2 Road Environment In addition to urban search and rescue the simulator has been applied to the DARPA Urban Challenge http www darpa mil grandchallenge This is supported by the ARDA map ARDA_RNDF txt and ARDA_MDFx txt files Formats for the Route Network Definition Files RNDF and Mission Description Files MDF are described on the DARPA web site Manual control of a robot does not require a RNDF or MDF Autonomous control or game scoring by DARPA rules will require those files The robot SnowStorm is equipped with sensors to drive autonomously in this environment 3 2 2 Sensor and effecter simulation Sensors are important to robot control Through checking the object s state or some calculation in the Unreal engine three kinds of sensor are simulated in US
80. art is shown below Sensor Odometry Sensor HumanMotion Sensor Sensor Encoder RFID Sensor Robot Camera Range Sensor RangeScanner Sensor Sonar Sensor IR Sensor SICK LMS200 IR Camera IR Scanner PB91 Irs Figure 528 Sensor Hierarchy Chart 14 2 2 Sensor Class The Sensor class is the ancestor of all the sensor classes It extends from the Item class which is the base class for all the items that can be mounted on the robot The Item class takes care of creating the item mounting itself on the robot providing a command response interface and preparing some information that will be helpful to the user The sensor class provides the basic interaction interface to send out data The details about Item classes are explained below Attributes var string ItemName the item s name var string ItemType the item s type var string ItemMount the mount base var vector myPosition the mounting position 147 var rotator myDirection the mounting direction var USARConverter converter the converter object used by USARSim to do unit and cooridniate conversion var K Vehicle Platform the item s robot platform Methods function SetName String iName set the item s name function Init String SName Actor parent vector position rotator direction KVehicle veh name mount mount the item function ConvertParam USARConverter converter tra
81. assname gt This is a very powerful command You can for example go with the spectator camera near a robot and type editactor class KRobot and the property window of that actor will open It s just like in UnrealEditor but this time you can change any parameter in real time in the simulation including physics Play with it There are a lot of other useful commands that work only in client mode follow this link http wiki beyondunreal com wiki Console_Commands After the Unreal client is started you can attach the viewpoint to any robot in the simulator Go to the Unreal client click left mouse button you will get the image 28 viewed from the robot To switch to next robot click left mouse button again To return back to the full viewpoint click the right mouse button When your viewpoint is attached to a robot you can press key C to get a viewpoint that looks over the robot Pressing C again will bring you back to the viewpoint of the robot TIP Left mouse button attaches your viewpoint to a robot Right mouse button returns your viewpoint to full viewpoint Pressing C let you switch viewpoint between robot s viewpoint and the overlook viewpoint Besides the step by step manual run of USARSim you can embed step and 2 into your application That is let your application start the Unreal server and client for you and then start itself The examples in the following section we will show you h
82. ates to maintain for example the following wheel of the P2DX robot the chassis of PER The state of the following wheel of P2DX is totally decided by the other two wheels This is not included in the general robot model So you need to maintain its state by yourself It s the same as the chassis of the PER PER s chassis is controlled by a differential that force the chassis s pitch angle to always be the average of the left and right wheel rocker angles To maintain the robot s state you need to override the Tick function In every Unreal tick you update the robot s state and you also need to explicitly or 157 implicitly call the Karma update state function KUpdateState You can use the code of P2DX as example to learn how to maintain your own state Lastly besides the three aspects mentioned above obviously you can do just about anything you want in your robot class 14 5 Build your controller The client server architecture makes it easy to build your own control client You only need to follow the communication protocol Since the protocol is line based you need to use the Win to determine when a message ends When you send out a command you need to add Win to inform USARSim that the command is finished In unreal engine a tick is the minimum time used for checking and updating states If you send commands at a higher frequency than the time interval between two ticks then the engine will only process t
83. ayerv and playernav you have to do this If you want to start player you have to do the following steps 1 start the USARSim server 2 adjust the host and port in your player configuration file to the USARSim server execute the player server player p2at cfg 4 execute playerv to see your sensor data 13 3 2 Device Drivers In this section we explain how the USARSim devices work and how to configure them 13 3 2 1 us_bot Synopsis The us_bot driver is the bridge between Gamebots and the USARSim devices It takes care the following tasks 1 Connect to the Unreal server specified by host and port 2 Spawn a robot of type bot at location pos 3 Collect and parse the robot s data from Gamebots 4 Provide the collected data to other USARSim devices and transfer these devices commands to Gamebots Interfaces Supported interfaces e simulation Required devices e None Configuration file options Name Type Default Meaning host string 127 0 0 1 The Unreal server name port int 3000 The Gamebots port number pos int int int 0 0 0 The initial spawn position 138 rot int int int 0 0 0 The initial spawn orientation bot string P2AT The robot type botname string The robot name 13 3 2 2 us_position Synopsis The us_position driver is used to control the robot s movement It gets the robot s steering type from USARSim and interp
84. ays be NauticVehicle for a nautic vehicle CONF message Name string string is the robot s name SteeringType string string gives the steering type of the nautic vehicle As of now this value can only be VariableDepth Mass float float is a the robot s mass in kilograms Example CONF Type NauticVehicle Name Submarine SteeringType Underwater Mass 50 0000 For an aerial vehicle the configuration message looks like still to be expanded CONF Type AerialVehicle Name string SteeringType string Mass float Where Type AerialVehicle The Type parameter is hard coded and will 62 always be Aerial Vehicle for an aerial vehicle CONF message Name string string is the robot s name SteeringType string string gives the steering type of the aerial vehicle As of now this value can only be Rotary Wing Mass float float is a the robot s mass in kilograms Example CONF Type AerialVehicle Name AirRobot SteeringType RotaryWing Mass 50 0000 For a sensor or effecter a configuration message looks like CONF Type string Name Value Name Value Where Type string specifies the sensor type string is the type name Name Value y is the name value pair that describes the feature of this sensor type Different sensor types have different name value pairs For detailed information please refer to section 10 about
85. bes the vehicle type It will be one of the following values GroundVehicle LeggedRobot NauticVehicle or AerialVehicle 51 Time float FrontSteer float RearSteer float SternPlaneAngle float RudderAngle float LightToggle bool LightIntensity int Battery int float is the UT time in seconds It starts from the time the UT server starts execution Note parameter only available for robots of GroundVehicle type Current front steer angle of the robot in radians Note parameter only available for robots of GroundVehicle type Current rear steer angle of the robot in radians Note parameter only available for robots of NauticVehicle type Current stern plane angle of the robot in radians Note parameter only available for robots of NauticVehicle type Current rudder angle of the robot in radians Indicate whether the headlight is turned on bool is true means on False value means off Light intensity of the headlight Right now it always is 100 int is the battery lifetime in second It s the total time remaining for the robot to run Example STA Type GroundVehicle Time 62 01 FrontSteer 0 0000 RearSteer 0 0000 LightToggle False LightIntensity 0 Battery 3564 A Mission Package state message reports the mission package links current properties Every mission package state includes a Name segment and se
86. ble commands at this level Try typing the following gt init gt dump gt mvl10 These commands perform the following functions init This initializes the robot platform It is necessary before any movement commands will be accepted dump This turns on a display of the robot s internal world model at this level myl This tells the robot to move to a position in local coordinates More information on running MOAST may be found at the MOAST website http moast sourceforge net 5 2 3 Pyro The Pyro plug in embeds the loading of the Unreal server client To start Pyro go to the pyro bin directory If you are using Windows OS execute the pyro py If you are on Linux run the shell file pyro After the Pyro interface is launched 31 1 2 3 4 5 6 Click the Simulator button and select USARSim py on the plugins simulators directory Select the arena you want to load on the plugins worlds USARSim py directory NOTE here USARSim py is a directory and not a file Pyro will automatically load the Unreal server and client for you Under linux OS the Unreal client is launched in another console Using Ctrl Alt F7 and Ctrl Alt F8 you can switch between the two consoles Click the Robot button and select the robot you want to add on the plugins robots USARBot directory You will see that the robot is added in the virtual environment You should now be able to control the robot using the Ro
87. bot menu To view the sensor data or camera state you can select the Service from the Load menu to load a service On the plugins services USARBot directory select the sensor you want to view You can also try to load a Brain to control the robot Click the Brain button and select a brain on the plugins brains For example you can select Slider py or Joystick py to control the robot You also can select BBWander py to let the robot wander in the arena For details about Pyro please read section 13 2 To switch among windows you can use Alt Tab To get control from UT2004 press Esc To pause the simulator switch to the Unreal client and then press Esc 5 2 4 Player Player is a device server So before you use Player you need to start USARSim Please follow step 1 and 2 in section 5 1 to launch USARSim As mentioned above it s hard to switch focus between the Unreal Client and other applications under Linux we recommend you launch USARSim on another machine After USARSim is started 1 2 3 You need to prepare the configuration file used for Player To learn how to prepare the configuration file please read the Player User Manual and section 13 3 1 A sample configuration file usarsim cfg is included in the player tar gz file You can simply copy this file to some place and use it to test Player Before going to the next step you need to change host name in the file to the host
88. canner LMS200 IMU sensor Odometry sensor Encoders 106 The P2DX specification is as follows Dimension Length x Width x Height 44 cm x 38 cm x 22 cm Wheel Diameter x Width 16 5 cm x 3 7 cm Weight 9kg Payload 20 kg Maximum translate speed 1800 mm s Maximum rotating speed 300 deg s Sonars positions are X mm Y mm Theta deg 155 115 50 155 115 50 190 80 30 210 25 10 210 25 10 190 80 30 155 115 50 115 130 90 12 3 2 Configure it It s the same as P2AT 12 4 ATRVJr 12 4 1 Introduction The ATRV Jr is a 4 wheel drive outdoor all terrain robot vehicle developed by iRobot In USARSim we use classname USARBot ATRVJr to represent the ATRV Jr a Real ATRVJr b Simulated ATRVJr Figure 28 ATRV Jr robot In summary the ATRV Jr has 107 e Four wheels e Differential steering In our simulation it is equipped with e PTZ camera e 17 Sonars 5 front 10 side 2 rear e Sick Laser Scanner LMS200 The ATRV Jr specification is as follows Dimension Length x Width x Height 77 5 cm x 62 2 cm x 55 cm Wheel Diameter x Width 33 cm x 10 cm guessed data Weight 50 kg Payload 25 kg Maximum translate speed 1000 mm s Maximum rotating Speed 120 deg s Sonars position are X mm Y mm Theta deg 334 95 104 39 30 y 340 41 49 91 15 y 347 06 0 0 340 41 49 91 15 334 95 104 39 30 230 23
89. ced USARSim drivers for Player 161 Version 1 6 from Erik Winter at Uppsala University Sweden Modified Version 1 6 from Stefan Markov amp Ravi Rathnam at International University Bremen Germany Version 2 0 from Stefan Stiene at University of Osnabriick System architecture Mission package concept from Stephen Balakirsky and Chris Scrapper at National Institute of Standards USA Unit and coordinate conversion idea from Chris Scrapper at National Institute of Standards USA Effecter concept from Chris Scrapper at National Institute of Standards USA The KDHinge from Marco Zaratti at University of Rome La Sapienza Italy Hummer by Ben Balaguer of NIST Stereo P2AT from Giuliano Polverari at University of Rome La Sapienza Italy Lisa Kurt2D Kurt3D from Stefan Stiene at University of Osnabriick Passarola from Ricardo Alc cer ricardoalcacer gmail com IST Instituto Superior T cnico Department of ISR Institute for Systems and Robotics Portugal Rugbot from Todor Stoyanov Jacobs University Breman Germany Kenaf from Kensuke Kurose of Tadokoro Laboratory Tohoku University Japan Sensors INU sensor from Stefan Markov and Ivan Delchev at International University Bremen Germany IR and Sonar sensor from Erik Winter at Uppsala University Sweden The first draft Encoder sensor was written by Andreas Niichter at University of Osnabriick Germany IR scanner from Giuliano Polverari at University of Rome La Sapien
90. ck to the application interface you must transform these coordinates and units back In USARSin all the conversions are implemented through the coordinate and unit converter class USARConverter To use your own coordinate and unit system in the application interface you need to built your converter class and configure USARSim to use it rather than the default USARConverter Scale is the ratio of the real object size to the corresponding size in the virtual world When you build a robot or world model you must follow the scale Otherwise you will get incorrectly scaled data in the application interface Of course you can have your own scale But you must change the converter class to make sure you get the correct data Tip While creating your own coordinate system and scale is possible it is not recommended Instead you should use the standards outlined in section 6 2 6 1 Coordinates Unreal Engine uses a left hand coordinate system Figure 17 The positive X axis extends in front of you and the positive Y axis is on your right hand The positive Z axis points straight up Z Y Figure 17 Left Hand Coordinate System In the application interface we use the right hand SAE J670 Vehicle Coordinate System Figure 18 The only difference from the Unreal Engine coordinate system is that the positive Z axis points straight down and not up 46 PITCH VELOCITY q A a are Yo A A LONGITUDINAL SS AXIS Nao VELOCITY u
91. configure it 10 2 Range Sensor 10 2 1 How the sensor works The range sensor is used to detect distance There are two types of range sensor in USARSim sonar and IR Basically the range sensor is simulated by emitting a line from the sensor position along the direction of the sensor in the Unreal world The first point met by the line is the hit point And the distance between the hit point and the sensor is the returned range value If the range is beyond the sensor s detection range the sensor will return the maximum detection range Before the data is sent back a random number is added to simulate random noise Then a distortion curve is used to interpolate the range data to simulate the real range sensor For the sonar sensor instead of emitting one line from the sensor it tries to emit several lines within its beam cone The shortest distance detected by the lines is the Currently this attribute is only used to calculate the robot s payload It will not affect the sensor s dynamic characteristic The real physical variable used in Unreal Engine is the Mass variable 78 value returned by the sonar sensor For IR sensor only one line is used However the line can cross through transparent materials glass 10 2 2 How to configure it We configure the range sensor in the USARBot ini file The sonar and IR sensor s configuration looks like USARBot SonarSensor HiddenSensor true bWithTimeStamp true We
92. configure the properties of the robots such as the battery life and the frequency of data transmission etc The robots are based on the real robots and they have different capabilities This section will introduce the robots one by one and explain how to configure them We control the robot mobility with one of the many DRIVE commands For details about the different DRIVE commands please go back to section 9 4 12 1 P2AT 12 1 1 Introduction The P2AT is a 4 wheel drive all terrain pioneer robot from ActivMedia Robotics LLC For more information visit the ActivMedia s website http www activrobots com In USARSim we use classname USARBot P2AT to represent the P2AT de NN a Real P2AT b Simulated P2AT Figure 26 P2AT robot In summary the P2AT has e Four wheels 103 e Skid steer In our simulation it is equipped with PTZ camera Front sonar ring Rear sonar ring Sick Laser Scanner LMS200 INS Odometry sensor RFID sensor The P2DX specification is as follows Dimensions Length x Width x Height 50 cm x 49 cm x 26 cm Wheel Diameter x Width 22 cm x 7 5 cm Weight 14 kg Payload 40 kg Maximum Translate Speed 700 mm s Maximum Rotating Speed 140 deg s Sonars positions are X mm Y mm Theta deg 145 130 90 185 115 50 220 80 30 240 25 10 240 25 10 220 80 30 185 115 50 145 130 90 145 130 90 185 115 13
93. control command Once you added the left right parameters interpretation your robot should be able to be controlled by Pyro As an example you can open the source code of P2AT to learn how it supports the DRIVE Left xxx Right xxx command The CAMERA command is another command used to interact with Pyro You also can learn how to interpret it in the P2AT uc file 2 Add your own commands Besides supporting the commands used by USARSim you also can add your own command As we mentioned before the commands come from Gamebots A robot connects with Gamebots through its controller whose class is USARRemotebot Every USARRemotebot is associated with a USARBotConnection that listens to a TCP IP socket and parses the incoming commands Once a new command is received USARBotConnection realizes it and gets the value in the command Then it sets the corresponding variable in USARRemotebot to the new value In your robot class you only need to check the USARRemotebot s variable to get the command data In summary to add a new command 1 Add a new variable in USARRemotebot to store the command s data 2 In USARBotConnection add your code into the ProcessAction function to interpret your command and store it in the USARRemotebots s variable 3 In your robot class check the USARRemoteBot s variable to get the command and do something you want 3 Maintain the robot s state by yourself Some robots may have special st
94. cs On Mapping under Unreal Wiki http wiki beyondunreal com wiki Topics_On_ Mapping lists all the topics involved in mapping 14 1 2 Special effects Most of the special effects are obtained by applying special materials Please read the UDN Material Tutorial http udn epicgames com Two MaterialTutorial to have a sense of what an Unreal material is The mask effect parts of material are either opaque or transparent is achieved by using textures with an alpha channel The gray level in the alpha channel indicates how transparent the corresponding pixel will be Alpha channel with grid bitmap will bring us the grid fender effect The glass effect is simulated by semi transparent material A texture with a gray alpha channel will give us a semi transparent effect Using shaded material we can get higher fidelity effects The mirror effect is obtained by using scripted texture The basic idea is to put a camera in the place you want to put the mirror and then render the picture from the camera into the place where the mirror is The idea comes from Angel Mapper s reflection tutorial http angelmapper com tutorials reflections htm The details about how to add a mirror can be found at the Security Camera Tutorial that is located at http angelmapper com tutorials securitycamera htm According to the author this approach doesn t work online To fix this shortcoming a customized CameraTextureClient named myCameraTextureClie
95. e Maximum rotational velocity 1 rad s 12 21 2 Configure it It s the same as P2AT 12 21 3 Extended USARSim command for Passarola robot DRIVE XZAngle floaf ThrustPropeller float TailPropeller float Where XZAngle float float is the rotation angle of the support thrust motors bars that make possible change the altitude of the robot i e up down If we use normalized values the value range is 100 to 100 and corresponds to the bar s minimum and maximum rotation angle respectively Otherwise the value is the absolute rotation angle in radians per second ThrustPropeller floa float is the module of the velocity vector to be applied by the front thrusters to move the robot in the XOZ plane i e forward backward and up down as the value of XZAngle If we use normalized values the value range is 100 to 100 and corresponds to the robot s minimum and maximum velocity respectively Otherwise the value is the absolute linear velocity in meters per second TailPropeller float float is the rotational velocity i e left right If we use normalized values the value range is 100 to 100 and corresponds to the robot s minimum and maximum rotational velocity respectively Otherwise the value is the absolute rotational velocity in meters per second 125 Normalized boo Indicates whether we are using normalized values or not The default value is False which means absolute values are used Example DR
96. e Server simulates a web camera How to run it is described in section 1 Send out a picture when the client connects with it 2 Wait for the acknowledgement from the client 3 If the current time is the sending time triggered in the specified frame rate then send out the next picture 4 Goto step 2 The server supports both raw pictures and jpeg pictures The image data format is ImageType 1 byte ImageSize 4 bytes ImageData n bytes ImageType The format of the image It can be 0 raw data 1 jpeg in super quality 2 jpeg in good quality 3 jpeg in normal quality 4 jpeg in averagequality 5 jpeg in bad quality ImageSize The total length of ImageData in bytes ImageData The actual data of the image For raw data the ImageData is width 2 bytes height 2 bytes RGB 1 byte 1 byte 1 byte data from left to right from top to bottom For jpeg the ImageData is the real jpeg data which can be decompressed by any jpeg decoders The image transfer protocol is very simple When the client gets the image it sends back an acknowledgement meesage OK in plain text to the image server 160 To use the image server you only need to follow this simple protocol and the image format As an example of how to use the image server the source code of SimpleUI is included in the USARSim package The SimpleUI uses the FreeImage http sourceforge net projects freeimage DLL to decode jpeg pictures 15 Informatio
97. e geometric model is very important You must let the X axis of the model point to the head and the Y axis point to the right An incorrect axis will bring you incorrect pitch yaw and roll angles NOTE Make sure the geometric model has the correct x axis and y axis This will affect the attitude data 14 4 2 Step2 Construct the robot 14 4 2 1 Create the part wheel class Here we create a wrapper class for our part or wheel geometry model The part class looks like class part_class_name extends KDPart defaultproperties properties where part_class_name is the name of your part class In defaultproperties we point the StaticMesh to your part s geometry model set the part s Weight Mass and the Kparams Karma parameters For details please refer the next section For the wheel class the only difference is that the class extends from USARTire not KDPart 14 4 2 2 Create the robot class First you need to create a robot class that extends the KRobot The class should look like class robot_class_name extends KRobot config USAR defaultproperties properties where robot_class_name is the name of your class 14 4 2 3 Prepare the attributes and objects used for your robot In the defaultproperties block of the class you can set the attributes of the robot The attributes are MotorTorque The default motor torque in Karma Units Default value is 20 MaxTorque The maximum motor torque Defau
98. e robot at this level of control are init Initialize the system halt Provide an orderly safe and recoverable halt of all systems abort Provide an immediate and safe halt of all systems Some systems may need to be reset to recover from an abort shutdown Turn off power down the systems arc lt file gt Drive the arcs given in the file lt file gt wp lt file gt Drive straight line segments between points given in the file lt file gt rotate lt theta gt Rotate the robot to angle theta vel lt v gt lt w gt Drive the vehicle at velocity v with rotational velocity w ct lt secs gt Set the system cycle time to lt secs gt pars lt 5 vmax gt Configure the vmax amax wmax alphamax and cut parameters that are used for path following debug Set the debug output level of the software at this level More information on the commands available at this level of the hierarchy as well as the other levels may be found at the MOAST website It should be noted that the shell programs are provided to demonstrate how to connect to the robot at various levels of control and to provide some simple user debugging One of the features of MOAST is that all of the control interfaces are brought out over standardized interfaces with NML Complete documentation on the available buffers is available on the MOAST website Programs running on Windows compiled with Microsoft Visual C under cygwin compiled with GNU tools
99. ed in USARSim are Opcodes Description Activate The activate opcode is used to turn on a process such as a wielder or a paint gun or place the effecter in an active state such arming a disrupter Animate The animate opcode is used for moving parts such as grippers or rollers 49 on a roller table Fire The fire opcode implies a discrete action such as firing a nail from a nail gun or discharging a disrupter It can also be used to initiate multiple firing through the use of the parameters Release The release opcode is used to drop an item into the environment or to detach a mount item Reset The reset opcode is used for refreshing the effecter to its initial state or reloading the effecter NOP The NOP is a general no operation command that is send in order to return the effecter to an idle state 9 Communication amp Control Messages and commands In this section we introduce how to communicate between USARSim and your application It will help you understand how to control the robots in the USARSim virtual environment 9 1 TCP IP socket As was mentioned before Gamebots is the bridge between Unreal and the controller It opens a TCP IP socket for communication purposes The IP address of the TCP IP socket is the IP address of the machine that runs the Unreal server The default port number of the socket is 3000 and the maximum allowed number of connections number is 16 To cha
100. ed robots the QRIO and ERS are also part of the simulation as well as a nautical vehicle Submarine and an aerial vehicle Helicopter Information about these robots can be found in section 12 3 2 4 Communications Simulation The purpose of the Wireless Communications Server is to act as a middle man for messages passed between the robots dropping messages and connections between robots when not realistically feasible using wireless communication It has been implemented in UnrealScript and is automatically started when starting a BotDeathMatch and hence a UsarDeathMatch 10 The server listens on a port for connections from robots sending command messages for registering listening and opening connections Once connections between robots are set up these are handled on different sockets TcpLinks allowing the server to listen for more commands and allowing multiple connections to be handled The opening of a connection or the closing of a connection while sending a message is decided using the path loss estimated using the Wall Attenuation Factor Model P d dBm P do dBm 10n log C WAP nW gt C 1 nWxWAF nW lt C do The default values for P do is 49 67 dBm dy is 2 m n is 1 09 C is 5 and WAF is 6 625 based on measurements taken in Research 1 at International University Bremen When the path loss reaches below 93 dBm the connections are closed the attempt at opening a connection fails 3 2 4 1 Commu
101. el radius 6 06cm 114 e Maximum wheel spin speed 6 2832 rad sec 12 11 2 Configuration See P2AT 12 12 Talon 12 12 1 Introduction The Talon is a lightweight tracked vehicle built by Foster Miller www foster miller com for missions ranging from reconnaissance and weapons delivery to rescue In USARSim we use classname USARBot Talon to represent this robot gt a Real Talon b Simulated Talon Figure 31 Talon Robot In summary a Talon has e Two tracks One arm with two joints Weight 34 kg Payload 45 kg In USARSim the Talon is equipped with e 4 fixed color cameras e One gripper with two fingers e One odometry sensor e One INU sensor The Talon specification is as follows e Dimension Length x Width x Height 91 17cm x 59 03cm x 36 54cm e Wheel radius 0 28cm e Maximum wheel spin speed 6 43 rad sec 12 12 2 Configure it It s the same as P2AT 115 12 13 QRIO 12 13 1 Introduction The QRIO is an artificially intelligent humanoid designed and manufactured by Sony Although the QRIO was to be marketed and sold by Sony as an entertainement robot development stopped in January 2006 In USARSim we use classname USARBot QRIO to represent this robot a Real QRIO b Simulated QRIO Figure 36 QRIO Robot In summary the QRIO has e Two Legs each of which has 6 joints Two Arms each of which has 3 joints Head made of 2 Joints Pan and Tilt Rotation Body made of two joints
102. er driverTable endif 3 Go to the server drivers directory add usarsim to SUBDIRS in the file Makefile am 4 Under the Player directory append the following line into acinclude m4 right after other PLAYER_ADD_DRIVER sentences PLAYER_ADD_DRIVER usarsim drivers usarsim yes 5 Add undef INCLUDE_USARSIM to config h in file 6 In the configure in file add the following line to AC_OUTPUT server drivers usarsim Makefile 7 Go back to the Player directory use the following commands to generate the configure file aclocal autoconf automake add missing 8 Follow the Player User Manual to compile and install Player To install Player 2 0 3 and our USARSim Player drivers 1 Download Player 2 0 3 from http sourceforge net project showfiles php group_id 42445 2 Restore Player on your machine 3 Restore player tar gz located in the Tools section of the USARSim file release area to your Player directory Please make sure all the files are put under the correct directory Execute the autogen_usarsim sh script This script patches some player files It needs autoreconf installed on your machine 4 Follow the Player User Manual to compile and install Player That is execute the following commands configure make make install If you want to install Player in another directory rather than usr local you need to use the command confi
103. erverPackages USARVictims Editor EditorEngine EditPackages USARBot EditPackages USARBotAPI EditPackages USARMisPkg EditPackages USARModels EditPackages USARVictims UnrealEd UnrealEdEngine Use this section only if you want USARSim packages to automatically load up when you start Unrealkd EditPackages USARBot EditPackages USARBotAPI EditPackages USARMisPkg EditPackages USARModels EditPackages USARVictims NOTE You need to modify ut2004 ini before you build your own models 14 1 Build your arena An arena is an Unreal map It includes geometric models and objects in the environment The objects can be obstacles such as bricks or victims that can move their bodies Before building your arena we must keep in mind that all the meshes must be static meshes Karma objects only works well with static meshes In addition static meshes can accelerate 3D graphic rendering NOTE All the meshes must be static mesh The Karma engine only works well with static meshes 142 When you build a new arena there are three things you may need to do 1 build the geometric model 2 simulate some special effects and 3 add objects such as obstacles and victims into the arena The three things are explained in the following sections 14 1 1 Geometric model We have two options for building a geometric model One is to import an existing model into Unreal The other
104. erwise the status will be Failed string is the name of the camera currently attached to viewport4 If viewport4 has been disabled this parameter will be Disabled If viewport4 has been attached to a non existent camera this parameter will be None string is the status of viewport4 after the SET command has been issued The status will be OK when the camera attached to viewport4 was successfully changed Otherwise the status will be Failed Example RES Time 56 68 Type Viewports Config SingleView Status OK Viewportl Camera Status OK Viewport2 Disabled Status OK Viewport3 Disabled Status OK Viewport4 Disabled Status OK After a SET Type Camera command the following response message will be issued RES Time float Type Camera Name string FOV float Status string Where Time float Type Camera Name string float is the timestamp in the virtual world when the message is sent out It is represented in seconds Camera is hard coded into this parameter string 1s the name of the camera that will be described by the next two parameters 1 e FOV and Status 66 FOV float float is the current field of view of the camera being described in radians The current field of view is the field of view after a SET Type Camera has been issued Status string string is the status for the camera s field of view af
105. es utx UT2004 Textures USARSim_Objects_Textures utx UT2004 Textures USARSim_LeggedRobots_Textures utx UT2004 ChangeLog UT2004 README 4 3 Install USARSim For non developers 1 Install Unreal Tournament You will get a folder which we will refer to as UT2004 for these instructions 2 Install the Official Unreal Tournament Patch V3369 which you can download at http data unrealtournament com UT2004 WinPatch3369 exe 3 Download the usarsim 2004 files in the download section of the USARSim project page http sourceforge net project showfiles php group_id 145394 4 Extract the files you just downloaded into the UT2004 folder 5 Compile USARSim by running the make bat script in the UT2004 System folder For developers 1 Install Unreal Tournament You will get a folder which we will refer to as UT2004 for these instructions 2 Install the Official Unreal Tournament Patch V3369 which you can download at http data unrealtournament com UT2004 WinPatch3369 exe 3 Temporarily rename the UT2004 folder to usarsim Put yourself in the folder above usarsim 22 4 Check out the latest source code snapshot from SourceForge For instructions on doing this go to http sourceforge net cvs group_id 145394 5 This will check out the USARSim source code snapshot into your usarsim folder merging it with what s already there 6 Change usarsim back to UT2004 7 Download the base files in the download
106. ether the tag is a single shot tag or a multi shot tag They are deployed by placing them in UnrealEd If the tag s id is set to 1 the default then the tag id will be set to a unique value automatically Other values for the id will not be changed If the tags are within the MaxRange 56 of the RFID sensor mounted on the robot then the server sends the following message to the client SEN Type RFIDTag Name string ID int Location float float float Where Name string string is the sensor name o Victim Sensor SEN Type VictSensor Name string PartName string Location x y z PartName string Location x y z Where Name string string is the sensor name PartName string string is the name of the victim part that was discovered by the sensor It can be one of 7 values Head Arm Hand Chest Pelvis Leg and Foot Please note that the sensor does not differentiate between real victim s part and false alarms It is up to the controller to perform this task Location x y z Relavite location based on the sensor s position and rotation of the victim part where x y and z are in meters Example SEN Type VictSensor Name VictimSensor PartName Leg Location 2 05 0 33 0 46 PartName Leg Location 2 51 0 44 0 39 o Human Motion Detection SEN Type HumanMotion Name string Prob float Where Name string string is
107. f the robot s chassis WheelBase float float is the wheel base in meters The wheel base defines the distance between two wheels along the width y axis of the robot s chassis Example GEO Type GroundVehicle Name ATRVJr Dimensions 0 7744 0 6318 0 5754 COG 0 0000 0 0000 0 1000 WheelRadius 0 1922 WheelSeparation 0 5120 Wheelbase 0 3880 For a legged robot the geometry information message looks like still to be expanded GEO Type LeggedRobot Name string Dimensions x y z COG x y z Where Type LeggedRobot The Type parameter is hard coded and will always be LeggedRobot for a legged 58 robot GEO message Name string string is the robot s name Dimensions x y z x defines the robot s length y defines the robot s width and z describes the robot s height Please note that these values are in meters COG x y z x y and z identify the position of the center of gravity in meters calculated from the chassis origin Example GEO Type LeggedRobot Name ERS Dimensions 0 3190 0 1800 0 2780 COG 0 0000 0 0000 0 0000 For a nautic vehicle the geometry information message looks like still to be expanded GEO Type NauticVehicle Name string Dimensions x y z COG x y z Where Type NauticVehicle The Type parameter is hard coded and will always be NauticVehicle for a nautic vehicle GEO message Name string
108. fps of the images 4 Click on Start Remember you must run image server on the same machine where UT server is running 3 2 5 3 Running the Image Server with USARSim in Client Mode If you plan to use USARSim in client mode then do the following 1 Start the image server 2 In the image server select UT Client Mode enter a map name and chose the format resolution and fps of the images 3 Click on Start 3 2 6 MultiView The USARSim client allows one to only see from the camera s of one robot at a time To overcome this limitation a special extension called MultiView was developed 3 2 6 1 Enabling a map for MultiView In order to enable the MultiView you have to add a special object into the map You must use the Unreal Editor to do that Open your map in the Unreal Editor As shown in Figure 9 we will use the Soccer map as an example 14 Figure 9 Soccer map in the unreal editor Click on the pawn button as illustrated in Figure 10 It will open the Actor Classes window File Edit View Brush Build Tools Help AO T Al Sal olle V Use Actor as Parent IV Placeable classes Only KActor KNActorFactory Keypoint Light MeshEffect NavigationPoint NetworkT rigger Note ONST eleportPad Redeemenwarhead USARRobot UnrealPawn Vehicle WorldController Pickup Figure 10 Click on the pawn button 1
109. ge axis in Figure 53 that the part can spin around Another is the steering and suspension axis Steering Axis in Figure 53 that the part can steer around and travel along A hinge joint connects two parts by one axis l Steering can be locked Chassis body Susp nsion travel SteeringAxis A Rolling Wheel body pa S Tin Je Aric Figure 539 Car wheel joint The part joint pair is a structure defined below struct JointPart Part var name PartName var class lt KActor gt PartClass var vector DrawScale3D Joint var class lt KConstraint gt JointClass var bool bSteeringLocked var bool bSuspensionLocked var float BrakeTorque var name Parent var vector ParentPos var vector ParentAxis var vector ParentAxis2 var vector SelfPos var vector SelfAxis var vector SelfAxis2 154 where PartName The name of the part PartClass The part s class name It can be Class USARBot KDPart or the tire s class name DrawScale3D The scale along X Y and Z axes of the static mesh Please note this only changes how the part looks look In unreal engine it still uses the original mesh for collision detection Please use it carefully JointClass The joint s class name It should be class KCarWheelJoint or class KDHinge bSteeringLocked Indicates whether steering is locked if we are using a car wheel joint bSuspensionLocked Indicates whether suspension is locked if we are using a
110. gure prefix lt your directory gt 27 5 Run the simulator 5 1 The steps to run the simulator As of version 2 2 the preferred method of running USARSim is to run in client only mode This may be accomplished by executing the following start path_to_bin_dir ut2004 map_name game USARBot USARDeathMatch spectatoronly 1 TimeLimit 0 quic kstart true ini usarsim ini where map_name is the name of the map for example DM USAR_ yellow the yellow arena Additional maps are available from the files section of the USARSim sourceforge repository under the Maps section TIP The files start_usar bat located in the UT2004 System directory can save you time in typing command line arguments Start the Controller After the Unreal server is started you can run your own application When you start USARSim in client mode only you can use these debug commands in the UT console showlog This command opens a window where the log file is shown in real time This is like having the DOS server window Actually it s better because in the DOS server window you cannot see client debug output only server related output In client mode with showlog command you can see all showdebug With this command UT will print some debug info directly on the client window One interesting thing is the Location of the camera With this information you can quickly choose new starting points location 250 editactor class lt cl
111. he robot s wheels in radians per second float is the maximum torque of the robot in unreal units float is the maximum steering angle for robot s front wheels in radians Please note that this value will be O for skid steered vehicles float is the maximum steering angle for robot s rear wheels in radians Please note that this value will be O for skid steered vehicles 61 Example CONF Type GroundVehicle Name ATRVJr SteeringType SkidSteered Mass 50 0000 MaxSpeed 2 0943 MaxTorque 60 0000 MaxFrontSteer 0 0000 MaxRearSteer 0 0000 For a legged robot the configuration message looks like still to be expanded CONF Type LeggedRobot Name string SteeringType string Mass float Where Type LeggedRobot The Type parameter is hard coded and will always be LeggedRobot for a leged robot CONF message Name string string is the robot s name SteeringType string string gives the number of legs of the legged robot As of now this value can only be TwoLegs or FourLegs Mass float float is a the robot s mass in kilograms Example CONF Type LeggedRobot Name QRIO SteeringType TwoLegs Mass 50 0000 For a nautic vehicle the configuration message looks like still to be expanded CONF Type NauticVehicle Name string SteeringType string Mass float Where Type NauticVehicle The Type parameter is hard coded and will alw
112. he absolute lateral velocity in meters per second RotationalVelocity float is the rotational velocity If we use float normalized values the value range is 100 to 100 and corresponds to the robot s minimum and maximum rotational velocity respectively Otherwise the value is the absolute rotational velocity in radians per second Normalized bool Indicates whether we are using normalized values or not The default value is False which means absolute values are used Example DRIVE AltitudeVelocity 1 will elevate the robot at a rate of 1 meters per second DRIVE Rotational Velocity 0 1 will rotate the robot at a rate of 0 1 radians per second DRIVE LinearVelocity 3 will make the robot go backward at a rate of 3 meters per second o DRIVE WheelNumber int WheelSpeed float WheelSteer float WheelNumber int WheelSpeed float WheelSteer float Where WheelNumber float is the number of an OmniDrive robot s wheel int as defined in the USARBot ini WheelSpeed float is the spin speed in rad s of the chosen wheel float WheelSteer float is the steer angle in rad of the chosen wheel float 71 Example DRIVE WheelNumber 0 WheelSpeed 3 14 WheelSteer 0 75 WheelNumber 1 WheelSpeed 3 14 WheelSteer 0 75 This command will turn both wheels by 0 75 rad while giving them different velocities If you use the LISA robot setup the robot rotates on place o DRI
113. he last command So please don t send your commands at very high frequency NOTE Don t send your command at a frequency higher than the engine s state update frequency If you want to do some image processing or include the video feedback on your own interface there are some technical details you may need to know As discussed in section 10 13 there are four ways to get use video feedback Except directly using Unreal Client as a separated window the other three approaches are discussed below 14 5 1 Embedding Unreal Client The idea is to attach the Unreal Client into your application Basically under windows this can be reached in 4 steps 1 Get the window handle of Unreal Client For example in C we can use CWnd m_AppWnd Find Window NULL Unreal Tournament 2004 2 Move and scale the Unreal Client to your desired region In C it may looks like m_AppWnd gt SetWindowPos this 60 40 400 300 NULL where this is the pointer of your application 3 Modify the Unreal Client s window style to let it look like a part of your application For example we use the following C code to remove the title bar and change the border to thick frame m_AppWnd gt ModifyStyle WS_CAPTION NULL SWP_DRAWFRAME m_AppWnd gt ModifyStyle WS_THICKFRAME NULL SWP_DRAWFRAME 4 Set your application to be Unreal Client s parent window In C we use m_AppWnd gt SetParent this where
114. he structure of sensor mounting is ItemClass The sensor class or the type of the sensor ItemName The name assigned to the sensor Parent The part the sensor will mount on Position The mounting position relative to parent s geometric center Direction The direction the sensor is facing relative to its parent Cameras The cameras mounted on the robot It uses the same structure of the sensor The first camera is the main camera And you use The CAMERA command to control it For other cameras please use SET and MISPKG command to control its FOV and direction Note When you control the camera make sure bAbsoluteCamera is set to the correct value 105 12 2 StereoP2AT 12 2 1 Introduction The StereoP2AT is a modification of the P2AT The specifications are the same as that of the P2AT except that the StereoP2AT adds stereo vision In USARSim we use classname USARBot StereoP2AT to represent this robot 12 2 2 Configure it It s the same as P2AT 12 3 P2DX 12 3 1 Introduction The P2DX is the 2 wheel drive pioneer robot from ActivMedia Robotics LLC For more information please visit ActivMedia s website http www activrobots com In USARSim we use classname USARBot P2DX to represent the P2DX a Real P2DX b Simulated P2DX Figure 27 P2DX robot In summary the P2DX has e Two wheels e Differential steering In our simulation it is equipped with PTZ camera Front sonar ring Sick Laser S
115. here ClassName robot_class robot_class is the class name of the 67 Name robot_name Location x y z Rotation r p y robot It can be USARBot ATRVJr USARBot Zerg USARBot P2AT USARBot P2DX USARBot Hummer and any other robots built by the user robot_name is the robot s name It can be any string you want If you omit this block USARSim will give the robot a name x y z is the start position of the robot in meters from the world origin For different arenas we need different positions The recommended positions are listed on Table 2 for the USAR arenas Recommended start locations for worlds are given in a text file that is included with the world download Worlds are available in the maps file release area on sourceforge r p y is the starting roll pitch and yaw of the robot in radians with North being 0 yaw Table 2 Recommended start position for the arenas Arena Recommended Start Position Yellow 4 5 1 9 1 8 Orange 1 2 2 3 1 16 Red 0 76 2 3 1 8 Example INIT ClassName USARBot P2DX Location 4 5 1 9 1 8 Name R1 will add a pioneer P2DX robot e Control the Robot There are seven kinds of control command The first kind controls the left and right side wheels of a skid steered robot The second kind controls the front and rear wheels of an Ackerman steered robot The third kind controls an underwater robot The fourth kind controls an aerial vehic
116. id 82 10 5 ON 82 10 5 1 How the sensor works ccccccccccccessssessccececececeesesseaecececeeseeensnseceeeeeesenenes 82 105 2 Howto Configure asno e 83 TOG INS SOS iia DA A E DA Vies 86 1037 Encoder Sensor iia aia risas 88 10 7 1 How the Sensor Works cccccccccccccessessssecececccecsesenseaeceeececeesesssseaeeeseeeeeenes 88 10 7 2 Howto Configure Tessa TEE ANE 88 1082 KOE Te aT KTDI E E ERE A arena ede tesa snore es 89 10 8 1 How the Sensor works ccccccccccccesssssssecececcceceesesseaeceeececeesessnseceeeseceeeenes 89 10 8 2 Howto Configure Tis sicccasageccivsssdentsenccanents aia 89 10 97 REID SOS ria iS a ee bapa a AE ea eh Boas ieee 90 10 9 1 How the Sensor Works ccccccccccccesessessecececcceceesesseaeceeeceesesenssseaeeeeecseeeees 90 O92 SHOW LO COMM OUTS Al io A e E A oa 90 10 9 3 Choosing the right SensingMode e cc eeeececeenceeeeececeeeeeceeeeeceeeeeesteeeesaes 91 10 9 4 Detecting RFID Tags queno dicas 92 10 9 5 Reading RFID Tags Memoria soeqnceds se acouageeeaa deecasesaeeeeetiaees 93 10 9 6 Writing RFID Tags Memory is 93 10 9 7 Erasing RFID Tags Memory icisiscccssscccsisassgeaecasescesashessaesavensceesntcceonanesanted 93 10 10 Victim and False Positive Sensor ooooooooncnccononnnonnnnnoncononannnnonononocnnononannnnnnnnos 94 10 10 1 How the sensor WOLKS cccccccccccccessessssececececeesesenseaeceseceeseeensnseaeeeeees 94 10 10 2 A e e a e NS 94 10 11 TONENE ESTEI iia
117. ief survey that may be found at http usarsim sourceforge net pages volunteer Survey html This survey will be used to aid us in getting further support for the development of this package It tells us who you are what you are using the package for and any suggested improvements Thanks 2 Introduction This manual is written for version 3 1 2 of USARSim The files may be found at the file release section of the USARSim web site http sourceforge net projects usarsim To install the base release please check the code out of cvs explained later in this document or download the USARSim full archive Maps and tools are also available on the website 2 1 Background Large scale coordination tasks in hazardous uncertain and time stressed environments are becoming increasingly important for fire rescue and military operations Substituting robots for people in the most dangerous activities could greatly reduce the risk to human life Because such emergencies are relatively rare and demand full focus on the immediate problems there is little opportunity to insert and experiment with robots 2 2 What is USARSim USARSim was designed as a high fidelity simulation of urban search and rescue USAR robots and environments intended as a research tool for the study of human robot interaction HRI and multirobot coordination Since its initial release it has been expanded to support many diverse environments including highway robots the D
118. ight 0 4 MaxRange 5 0 BeamAngle 0 3491 Noise 0 05 OutputCurve Points InVal 0 000000 Out V al 0 000000 In Val 5 000000 OutVal 5 000000 USARBot IRSensor HiddenSensor true bWithTimeStamp true MaxRange 5 0 Noise 0 05 OutputCurve Points InVal 0 000000 Out V al 0 000000 In Val 5 000000 OutVal 5 000000 Where HiddenSensor Boolean value is used to indicate whether the sensor will be visually shown in the simulator Setting it to true will hide the sensor We recommend setting it to true if it is not necessary to show the sensor When you want to confirm if the sensor is placed in the correct position and has the correct direction you can temporarily set it to false bWithTimeStamp Indicates whether the timestamp in the Unreal Engine will be associated with the sensor data Weight The weight of the sensor in kg MaxRange The maximum distance that can be detected in meters BeamAngle The sensor s detection cone in radians Noise The relative random noise amplitude With the noise the data will be data data random noise data where random noise returns a value between noise and noise OutputCurve The distortion curve It is constructed by a series of points that describe the curve 79 10 3 Range Scanner Sensor 10 3 1 How the sensor works The range scanner sensor is very similar to the range sensor In USARSim we treat the range scanner sensor as a series of range sensors The data is obtained by
119. im robot where this device is located NOTE The camera should use absolute pose control 13 3 2 8 Known bugs 1 2 There is frequently the Read Timeout while trying to read from server error in us_bot cc But it seems like this doesn t affect anything Sometimes the us_position device is not able to get the robot configuration from USARSim If that s the fact you can t control the robot from player or you even get a can t subscribe to position2d device error and player shuts down If you subscribe to a laser or a sonar device and the robot is not in 0 0 0 orientation the sensors point in the wrong direction This is because 141 USARSim GEO message supports the sensor orientation in a global coordinate system and not in a robot centered coordinate system 14 Advanced User This section is for advanced users who want to build their own additions to the simulator We assume the user already has programming experience or 3D modeling experience and robot background Before we start this section we need to change the ut2004 ini file found in the Unreal system directory Adding the following lines to the corresponding sections in ut2004 ini will let the Unreal engine recognize our own model With this modification we can compile and use our models in Unreal Editor Engine GameEngine ServerPackages USARBot ServerPackages USARBotAPI ServerPackages USARMisPkg ServerPackages USARModels S
120. imum spin angle the sensor can recognize in radians Noise The relative random noise amplitude With the noise the data will be data data random noise data where random noise returns a value between noise and noise NOTE Mounting the sensor on the front or back side of a wheel will give us a different count sign 10 8 Touch Sensor 10 8 1 How the sensor works We use the same method used by the range sensor to simulate the touch sensor Every touch sensor is treated as a button We emit several lines from the button s surface to detect the objects in front of the sensor When one object is close enough to the sensor less than 4 7mm the sensor will send out a touch signal 10 8 2 How to configure it We configure the sensor in the USARBot ini file The configuration looks like USARBot TouchSensor HiddenSensor true bWithTimeStamp False Weight 0 4 Diameter 0 01 Where HiddenSensor Boolean value is used to indicate whether the sensor will be visually shown in the simulator Setting it to true will hide the sensor We recommend setting it to true if it s not necessary to show the sensor When you want to confirm if the sensor is placed in the correct position and has the correct direction you can temporarily set it to false bWithTimeStamp Indicates whether the timestamp in the Unreal Engine will be associated with the sensor data Weight The weight of the sensor in kg Diameter The diameter of the senso
121. irements For the controller we recommend MOAST which is fully integrated with USARSim In addition Pyro out of date or Player may be used The Pyro plug in requires Pyro 2 2 1 and the Player USARSim drivers require Player 1 4rc2 or higher 4 2 Install UT2004 4 2 1 Windows Please note if you have a previous version of USARSim installed it must be uninstalled Users have three options to remove previous USARSim versions Please note that UT20024 refers to the Unreal Tournament main folder a Use the Uninstaller USARSim Full version 3 00 or higher only 1 Run the Uninstall USARSimFull exe file located in the UT2004 directory 20 b Re Install Unreal Tournament 1 Uninstall Unreal Tournament by going to Start gt Programs gt Unreal Tournament 2004 gt Uninstall Unreal Tournament 2004 2 Manually delete the 9UT2004 folder from your computer 3 Install Unreal Tournament 4 Install the Official Unreal Tournament Patch V3369 which you can download at http data unrealtournament com UT2004 WinPatch3369 exe c Manually Delete Files 1 Manually delete the following folders if they exist UT2004 Doc UT2004 Tools UT2004 BotAPh UT2004 USARBot UT2004 USARBotAPI UT2004 USARMisPkeg UT2004 USARModels UT2004 USAR Victims 2 Manually delete the following files if they exist UT2004 Animations UDN_CharacterModels_K ukx UT2004 Sounds SEER Voices uax UT2004 StaticMeshes FreiburgRescue us
122. is a toy based robot which was first turned into a robot platform named Lurker by the team Rescue Robots Freiburg They used the modified version in the Rescue Robot League during the RoboCup 2005 competition The Tarantula model which is now part of the USARSim package was originally developed at the University of Freiburg and has been further improved and merged into USARSim by the University of Pittsburgh In USARSim we use classname USARBot Tarantula to represent this robot Y a Real Tarantula b Simulated Tarantula Figure 34 Tarantula Model in USARSim In summary the tarantula has e Four flippers each of which can be controlled independently The tarantula specification is as follows e Dimension Length x Width x Height 0 6778m x 0 444m x 0 1406m e Maximum wheel spin speed 3 1416 rad sec The most significant difference of the Tarantula as compared to conventional robot platforms is the flippers which endow the robot with considerable mobility The flippers control commands are e Front flipper MULTIDRIVE FRFlipper float FLflipper float e Rear flipper MULTIDRIVE RRFlipper float RLFlipper float For example MULTIDRIVE FRFlipper 1 57 FLFlipper 1 57 will flip the front flippers towards 90 degree Sensors cameras and effectors can be attached to the Tarantula model in the same way as to any other robot as for example a P2AT 113 12 10 2Configuration In addition to standard configura
123. is always 18 8cm above ground actually it may vary from 18 to 19 cm because the value is rounded to the nearest cm it s easy to write the following table other robots may be added if needed P2AT 26 cm 7cm P2DX 20 cm 1 cm Zerg 6 cm 13 cm Talon 13 cm 6 cm ATRVjr 39 cm 20 cm e Position above ground is the Z spawning point for that robot assuming that the ground is at Z 0 e Delta Z is the value to be added to the Z returned by GETSTARTPOSES command 5 3 6 1 Example GETSTARTPOSES returns these are the real values from the map you have seen in the flash movie NFO StartPoses 4 Start1 1 79 0 70 0 19 0 00 0 00 0 00 Start2 4 10 0 19 1 21 0 00 0 00 0 00 Start3 4 09 1 79 1 72 0 00 0 00 0 00 Start4 0 96 1 86 2 24 0 00 0 11 0 51 If I want to spawn a P2AT at the Start4 position I will use as location 0 96 1 86 2 31 because Z 2 24 0 07 m If I want to spawn a Zerg at the same location I will use 0 96 1 86 2 11 because Z 2 24 0 13 m 6 Coordinates Units and Scale In USARSim there are two kinds of coordinates and units One is the coordinate and unit system used in the Unreal Engine Another is the coordinate and unit system used by the user applications interface If you are programming in Unreal Engine it s your responsibility to convert from the application interface coordinate and unit system to the unreal engine coordinate and unit system When you want to 45 send data ba
124. is an Ackerman steered vehicle based on the Mini Cooper It was added to USARSim as an additional vehicle for outdoor scenes a Real Cooper b Simulated Cooper Figure 32 Cooper Vehicle In summary the Cooper has 111 e Four wheels The two front wheels are powered e Single Ackerman steering front two wheels are Ackerman steered The Cooper specification is as follows e Maximum translate speed 134 172 kph 83 37mph e Maximum wheel spin speed 100 rad s 12 8 2 Configure it It s the same as P2AT 12 9 Submarine 12 9 1 Introduction The submarine is an underwater robot It is a standard submarine which was not based on any particular real models In USARSim we use classname USARBot Submarine to represent this vehicle Figure 33 Submarine in Simulation In summary the submarine has e A propeller a rudder and a stern plane In our simulation the submarine has e A pan tilt camera that can take a 360 degree panorama e A sonar sensor located on the belly of the submarine pointing down The submarine specification is as follows Dimension Length x Width x Height 6 5364m x 1 1428m x 1 9929m Rudder Area 0 3378m Maximum Rudder Angle 0 4363 radians Stern Plane Area 0 2152m Maximum Stern Plane Angle 0 4363 radians Propeller s Pitch 0 3048m Maximum Propeller Spin Speed 6 28 rad s 112 12 9 2 Configure it It s the same as P2AT 12 10 Tarantula 12 10 1 Introduction The Tarantula
125. is to build the model by hand in Unreal After building the model we need to transfer it into a static mesh To facilitate users building their own arenas we modeled all the parts used for building the NIST arenas The model packages are located in the file ut2004 StaticMeshes NIST usx and ut2004 StaticMeshes USAR_Meshes usx olf Wy f Figure 50 Some NIST facilities 14 1 1 1 Import an existing model The basic idea of importing a model is to convert your model into a format that Unreal Editor can read in The file formats that are supported by the Unreal engine are e ASC A 3D graphics file created from 3D Studio Max e ASE Short for ASCII Scene Exporter e DXF 3D graphic image file originally created by AutoDesk which stores 3D scenes and models e LWO Is from LightWave model program e T3D Is a text file that holds a text list of Unreal map objects Details about how to import a 3D model are described in the document 143 UDN Converting CAD data into Unreal http udn epicgames com Two CADtoUnreal 14 1 1 2 Build it with Unreal Editor Unreal Editor is a nice 3D authoring tool There are two websites you may need to visit if you want to learn how to build a map with Unreal Editor UDN Unreal Developer Network http udn epicgames com Unreal Wiki http wiki beyondunreal com wiki The Basics category in UDN contains documents with all of the details of modeling with the Unreal Editor And the Topi
126. its properties to get a new type of sensor In USARSim all of the sensors can add noise and apply distortion to their data except for the state sensor and robot camera This noise is applied to the output values reported by the sensor and not to the control values For example the angle between range scans for a scanning laser is always correct and a sensor will always point to the location that is commanded The distortion curve is a function such that output_data distortion input_data and the function itself is defined by a series of x y points connected by straight line segments If the input_data is outside of the defined range zero will be returned For the sensor output changing parameters will give different quality sensor data Every sensor has a Weight attribute associated to it Besides this it s also possible to associate a timestamp to the sensor data To do this we can configure the sensor with the variable bWithTimeStamp set to true in the configuration file In this section we will explain how the sensors work and how to configure them To learn how to build your own sensor please read section 14 2 10 1 State Sensor 10 1 1 How the sensor works The State sensor reports the robot s state Basically it just checks the robot s state in the Unreal engine and then sends it out Please go to section 7 3 for a detailed look at the state message 10 1 2 How to configure it None We do not need to
127. kgClass Class USARMisPkg OmniCamPillar Cameras ItemClass classUSARBot RobotCamera IltemName Camera Parent OmniCamPillar_Link4 Position Y 0 0 X 0 0 Z 0 09 Direction Y 1 5707964 Z 3 1415927 X 0 0 CamTexActors ItemClass classUSARBot USAREnmitter ItemName OmniCamEmitt er Parent OmniCamPillar_Link4 Position Y 0 0 X 0 0 Z 0 0 Direction Y 0 0 Z 0 0 X 0 0 The pillar consist of a bottom link multiple center links and a top link To adjust pillar height center links can be added or removed in the USARMisPkg ini file 10 15 3 How to configure it The configuration of a standard robot camera in the USARBot ini is used USARBot RobotCamera CameraDefFov 0 7854 CameraMinFov 0 3491 CameraMaxFov 2 0944 T Schmits and A Visser 2008 An Omnidirectional Camera Simulation for the USARSim World Proceedings CD of the 12th RoboCup International Symposium Suzhou China To be published in the Lecture Notes in Artificial Intelligence Springer Verlag Berlin Heidelberg Germany 11 Effecters 11 1 Gripper 11 1 1 How the effecter works A gripper constructed by a gripper base a left finger and a right finger The gripper base is the part where the effect is mounted The left and right finger can be explicitly defined by the user or automatically found by USARSim through iterating all the parts connected to the gripper base This effecter controls the left and right fingers joints to open o
128. ks at a parabolic shaped texture which is a monitor connected to a security camera hidden behind the texture The security camera is located at the center of projection of the parabolic texture and looks through the texture in front of it The UsarSim camera looks at the other site of the texture and captures the image displayed on the texture as if it is a reflection Care has been taken that the image of the security camera correctly distorted on the parabolic texture to get the perspective of an omnidirectional mirror 10 15 2 How to mount it To mount an omnidirectional camera you need three elements MisPkgs PkgName OmniCamera Location Y 0 0 X 0 0 Z 0 0 PkgClass Class US ARMisPkg OmniCamera Cameras ItemClass class USARBot RobotCamera IltemName Camera Parent OmniCamera_Link1 Position Y 0 0 X 0 0 Z 0 09 Direction Y 1 5707964 Z 3 1415927 X 0 0 CamTexActors ItemClass classUSARBot USAREnmitter ItemName OmniCamEmitt er Parent OmniCamera_Link1 Position Y 0 0 X 0 0 Z 0 0 99 Direction Y 0 0 Z 0 0 X 0 0 The UsarEmitter class is used for the security camera The OmniCamera Mission Package consists of the OmniCamBase and the OmniCamMirror models fixed to each other The camera can be mounted on a pillar to give it more height Use the following configuration to mount the camera on a pillar MisPkgs PkgName OmniCamPillar Location Y 0 0 X 0 0 Z 0 0 P
129. lane VerticalFOV The field of view in the vertical direction x z plane in radians VerticalStep The amount of radians between two traces in the vertical 94 bWithTimeStamp bShowResults Mean Sigma 10 11 Sound sensor direction x z plane Indicates whether the timestamp in the Unreal Engine will be associated with the sensor data Indicates whether the graphical interface sgould be displayed Useful for testing purposes but will be set to false in most cases Sets the mean used by the Gaussian random number generator used to produce noise in the sensor Sets the standard deviation used by the Gaussian random number generator used to produce noise in the sensor 10 11 1How the sensor works The Sound sensor detects sound sources in USARSim The sound sensor finds all the sound sources and calculates the source that is the loudest at the robot s location The loudness decreases with the square of the distance Currently the only available sound sources are victims Please refer to section 14 1 3 about how to associate a sound source to a victim 10 11 2How to configure it The sound sensor configuration in the USARBot ini file looks like USARBot SoundSensor HiddenSensor True Weight 0 4 Noise 0 05 OutputCurve Points In Val 0 000000 Out V al 0 000000 In V al 1000 0000 00 OutVal 1000 000000 Where HiddenSensor Weight Noise OutputCurve Boolean value is used to indicate whethe
130. le The fifth kind controls the wheels of an OmniDrive robot The sixth kind controls a specified joint of the robot The seventh kind controls the angle of multiple joints of a robot which is convenient for flipper leg and arm control o DRIVE Left float Right float Normalized bool Light bool Flip bool Where Left float float is spin speed for the left side wheels If we are using normalized values the value range is 100 to 100 and corresponds to the robot s minimum and maximum spin speed If we use absolute values the value will be the real spin speed in radians per second Right float Same as above except the values affect the right side 68 Normalized bool Light bool Flip bool wheels Indicates whether we are using normalized values or not The default value is False which means absolute values are used to control wheel spin speed bool is whether turn on or turn off the headlight The possible values are True False If a robot rolls over or otherwise tips off of its wheels this will right the robot If bool is True this command will flip the robot to its wheels down position Example DRIVE Left 1 0 Right 1 0 will drive the robot moving forward with spin seed 1 radian per second DRIVE Left 1 0 Right 1 0 will turn the robot to left side DRIVE Light true will turn on the headlight DRIVE Flip true will flip the robot o DRIVE Speed fl
131. le The speeds of the rollers are synchronously controlled by the effecter which enables the user to adjust the speed of the rollers to move an item along the surface of the table The directions of the rollers are determined by the sign of the parameter i e a positive value will spin the rollers the positive direction along the positive X axis and a negative value will cause the rollers to spin in the opposite direction Table 8 Roller Table opcodes Opcodes Parameters Status Animate Float that defines the speed OK successful and direction that the Failed to set the speed of rollers rollers will move in radians Error may be caused by improper per second arguments 11 3 2 How to configure it The Roller Table can be configured in the USARBot ini file The Roller Table section of this initialization file looks like USARBot RollerTable MaxSpeed 6 0 102 Where MaxSpeed Determines the maximum angular velocity that the rollers can spin in either direction 11 4 11 5 Headlight The headlight should also be an effecter Right now it s NOT an effecter because we simulate it by extending it from an Unreal class DynamicProjector In the future we should try to fix this problem 12 Robots All robots in USARSim have a chassis multiple wheels sensors and effecters The robots are configurable You can specify which sensors effecters are used and where they are mounted You also can
132. lt value The proportional gap used by a hinge 364 0 joint The proportional gap used for steering 1000 0 speed control The torque applied to the steering 1000 0 The steering speed 15000 0 Stiffness of suspension springs 150 0 Damping of suspension 15 0 The highest offset from the suspension 1 0 center in Karma scale which is 1 50th of Unreal scale The lowest offset from the suspension 1 0 center in Karma scale which is 1 50th of Unreal scale Roll friction of the tire 15 0 Lateral friction of the tire 15 0 Maximum first order force velocity 0 06 slip in tire direction Maximum first order force velocity 0 06 slip in sideway direction The minimum slip in both directions 0 001 The amount of slip per unit of velocity 0 0005 The softness of the tire 0 0 The stickyness of the tire 0 0 The bouncyness of the tire 0 0 153 TIPS Low TireSlipRate and high friction give the tire high climbing capability 14 4 2 4 Connect the parts wheels After we set up all the attributes and classes we can use the part joint pairs to connect the chassis and parts In the part joint pair we define the part and how it is connected to another part through a joint Currently we support two kinds of joints the car wheel joint that is used to connect a wheel to the robot and the hinge joint that is used to link any parts together A car wheel joint connects two parts by two axes One is the spin axis hin
133. lt value is 60 The control torque will be cut to this value if it s larger than MaxTorque MotorSpeed The default motor speed Default value is 0 1745 radians second Weight The weight of the chassis in kg Please note the value is 152 ChassisMass StaticMesh DrawScale DrawScale3D KParams ConverterClass used only for description purpose The real value that affects the physical characteristic is ChassisMass The mass of the chassis in Karma Units Default value is 1 0 The static mesh for the chassis The format looks like StaticMesh your_mesh_name The scale of the static mesh Default is 0 3 The scale in X Y and Z axes The Karma physical parameters of the chassis It s a KarmaParams object For details please read the UDN KarmaReference http udn epicgames com Two KarmaReference The class used by the robot for units and coordinates conversion By default it s USARBot USARConverter Besides these properties you also can set the joints and tire parameters for the robot These parameters will affect all the joints and tires Usually you needn t change them In case you want to change them we list all the parameters below Name HingePropGap SteerPropGap SteerTorque SteerSpeed SuspStiffness SuspDamping SuspHighLimit SuspLowLimit TireRollFriction TireLateralFriction TireRollSlip TireLateralSlip TireMinSlip TireSlipRate TireSoftness TireAdhesion TireRestitution Description Defau
134. n SI units and all the data you send to USARSim should also be in SI units 7 Mission Package The mission package concept is used throughout this manual Therefore we introduce this concept before we explain any other detailed information about USARSim A mission package is a virtual parts and joints container used for organizational and control purposes It is constructed of connected parts that work together to fulfill a behavior which is not related to the robot s mobility For example the camera s pan and tilt parts work together to adjust the camera s pose as shown in Figure 19 A part is connected to another by attaching a mount to a joint Every part and mount has its own local coordinate frame The joint control changes the relationship between the part s coordinate frame and the mount s coordinate frame In general a mission package is described as a part set and every part has 0 N N gt 0 mount locations The parts connect to each other on the mounts through joints Figure 20 is a more general example that depicts an arm mission package Pan gt Tilt De mz mx C 7 era CX CZ Mount mz Figure 19 Mission package and its coordinates 48 PartA Sensor Mount PartB Arm Part SM2 Joint JM2 Hard Mount Joint Mount lt O a p o E Q o El Figure 20 Arm mission package When building a robot we define a mission package as a list of part joint pairs E
135. n for Gamers Some public UT2004 game servers running the AntiTCC anti cheat service may be set up to automatically ban users who have Hook dll installed as it or at least files like it can be used for aimbots programs used by cheats to improve aim USARSim uses this file to hook into UT2004 s framebuffer to get images for the simulated camera and it can be found in the lt UT2004directory gt System directory alongside UT2004 exe The best way around this is to temporarily move hook dll to somewhere outside the UT2004 directory whilst playing on public servers Saving the following 3 lines to something like playUT bat putting it in lt UT2004directory gt System and making a play link that points to it is one way around this problem move hook dll UT2004 exe move hook dll 16 Bug report Please use the sourceforge message forums to report bugs in USARSim This forum may be found at http sourceforge net tracker group_id 145394 amp atid 761824 17 Contributors The primary software author e Jijun Wang at University of Pittsburgh USA e Major rewrites and contributions by Ben Balaguer of NIST USA The primary manual authors e Jijun Wang at University of Pittsburgh USA e Stephen Balakirsky of NIST USA Management and Coordination e Stephen Balakirsky at National Institute of Standards USA e Michael Lewis at University of Pittsburgh USA e Stefano Carpin at University of California Mer
136. n in the Mode group To return to the command mode click the right button of the mouse For joystick If you have set joystick enabled in Unreal you need to disable it so the system will not be confused The ways of using joystick are e Pushing the joystick forward backward will move the robot forward backward 29 e Pushing the joystick to left right side will turn the robot to left right e Pushing POV button up down will tilt the camera e Pushing POV button left right will pan the camera For keyboard mouse Since the interface and Unreal share the keyboard and mouse when you control the robot you MUST let the interface be active Otherwise the interface cannot get the input from the keyboard and the mouse To control the robot e Up Down Arrow key moves the robot forward backward e Left right Arrow key turns the robot to left right e Move mouse up down to tilt the camera e Move mouse left right to pan the camera 5 2 2 MOAST MOAST is configured for a particular simulation environment through the use of its initialization file Control of which MOAST modules will be run is controlled through a run script 5 2 2 1 Configuration Before starting MOAST the Unreal server must be running The operation of MOAST is controlled through the moast ini file located in the dev etc directory under the MOAST home directory For the novice user there are only two entries of interest in this file The first is the entry HOS
137. n is automatically set to accommodate for the maximum number of cameras In other words if a robot has zero or one camera it will automatically start in SingleView otherwise it will automatically start in QuadView Since a robot might have more than four cameras and users might want to cycle through all the cameras using SingleView USARSim gives the possibility of configuring the viewports as follows SET Type Viewports Config int string Viewportl string Viewport2 string Viewport3 string Viewport4 string Where Type Viewports Viewports is a hard coded parameter Config int string This parameter defines the viewports configuration It can be one of two values 73 0 or Single View for one viewport 1 or QuadView for four viewports Viewportl string string is the name of the camera that will be attached to viewport 1 Optionally string can be set to Disable to disable viewport 1 Viewport is the only viewport in Single View and the Top Left viewport in QuadView Viewport2 string string is the name of the camera that will be attached to viewport 2 Optionally string can be set to Disable to disable viewport 2 Viewport 2 is the Top Right viewport in QuadView Viewport3 string string is the name of the camera that will be attached to viewport 3 Optionally string can be set to Disable to disable view
138. nal 0 by default and indicates the control mode 0 means angle control 1 means speed control and 2 means torque control 75 Example MISPKG Name CameraPanTilt Link 1 Value 1 5 Link 2 Value 0 MISPKG Name TalonArm Link 1 Value 0 70 Link 3 Value 0 1 Order 1 TIPS Pan tilt mission packages enable the use of multiple cameras with independent control on each of them e Control a sensor effecter This command is used to send a command to a sensor or effecter The command looks like SET Type string Name string Opcode string Params value Where Type string string is the sensor or effecter s type Name string string is the sensor or effecter s name Opcode string string is the operation code Different sensors or effecters Valid opcodes are defined in Section 8 Params value value are the parameters associated with the operation command Example SET Type Odometry Name Odol Opcode RESET Params 1 e Control the Trace USARSim has the capability of tracing the path that a robot takes Colored Navigation Points are dropped into the world as the robot travels effectively tracing the robot s path Tracing can be used with the following command Trace On bool Interval float Color int string Where On bool bool tells USARSim to start stop tracing the robot s path Default value is false Interval float float
139. nd sends commands to control the robot The client server architecture of Unreal makes it possible to add multiple robots into the simulator However since every robot uses a socket to communicate for every robot the controller must create a connection for it The Mobility Open Architecture Simulation and Tools MOAST framework is designed to allow researchers to concentrate their efforts in their area of expertise To accomplish this the framework provides a hierarchical modular set of controllers interfaces and tools The controllers conform to the hierarchical 4 D RCS Reference Model Architecture Albus 2000 and provide behavior generation world modeling and sensor processing The hierarchy supports control ranging from low level servo control to high level robot team control To utilize this framework experimental code connects to one or more of the standardized interfaces to obtain data from the robot s and exert control MOAST is developed to fully integrate with the USARSim simulation system More information about MOAST is located at http moast sourceforge net A detailed explanation of the MOAST interfaces may be found in section 13 1 Besides MOAST USARSim also supports two other popular robot controllers Pyro out of date and Player The Pyro plug in included in USARSim allows the use of Pyro to control the robot in the simulator Pyro http pyrorobotics org is a Python library environment GUI and low level driver
140. nge these parameters we can go to the BotAPLini file in the Unreal system directory The section BotAPI BotServer of BotAPI ini looks like BotAPI BotServer ListenPort 3000 MaxConnections 16 Where ListenPort is the port number of the socket MaxConnections is the maximum number of connections It also decides the maximum number of robots you can add into the virtual world You can change or add if you cannot find the parameters in the INI file the parameters to the values that you want 9 2 The protocol NOTE A Name or String as referred to in this manual is defined as any consecutive collection of non white space printable ASCII characters with no delimiters such as single or double quotes This is ASCII characters 33 through 126 More information may be found at http udn epicgames com Two UnrealScriptReference Variables The communication protocol is the Gamebots protocol All of the data messages and commands follow the format 50 data_type segmentl segment2 where data_type specify the type of the data It is upper case characters Such as INIT STA SEN DRIVE etc segment is a list of name value pairs Name and value are separated by a space For example for Location 4 5 1 9 1 8 the name is Location the value is 4 5 1 9 1 8 For the segment Name Left Range 1 5 the names are Name and Range the values are Left and 1 5 A message
141. nication Base Station A communication base station is provided with USARSim and should be used as a relay point for your robots The communication base station is added into a world by issuing the following command INIT ClassName USARBot ComStation Location x y z If you are unfamiliar with the INIT command please read section 7 4 3 2 4 2 Configuration Several options can be changed using the USARComServer ini file For the USARBotAPI ComServer class ListenPort the port the server listens on bDebug Boolean flag indicating the printing of log messages For the USARBotAPI ComConnection class Jacobs University Bremen as of Spring 2007 11 bDebug Boolean flag indicating the printing of log messages ePdo The signal strength at a reference distance do in dBm eDo The reference distance do eN The log factor n eCutoff The cutoff signal strength in dBm eMaxObs The maximum number of obstacles C in the formula above eAttenFac The signal attenuation for each obstacle WAF in the formula above For the USARBotAPI ComLink class bDebug Boolean flag indicating the printing of log messages 3 2 4 3 Message formats Registering REGISTER RobotName IPAddress The RobotName should be the same as the one the robot was created with in the simulation The PAddress is the IP Address of the computer on which the robot controller is running The REGISTER word is not case sensitive Listening LISTEN RobotName Po
142. nsfer the item s parameters units and coordinates to Unreal units and coordinates function string Set String opcode String args the interface of the SET command function bool isType String type return whether the item s type matches the specified type function bool isName String name return whether the item s name matches the specified name The new variables and functions introduced in the Sensor class are Attributes var config bool HiddenSensor variable that indicates whether to show the sensor in Unreal var config InterpCurve OutputCurve the distortion curve var config float Noise the random noise amount Methods function String GetHead the interface that sends sensor data HEAD to the robot It s usually something like SEN Type xxx function String GetData the interface that sends sensor data to the robot For example it can be Name xxx Pose x y theta function String GetGeoHead the interface that sends the sensor s geometric information HEAD to the robot function String GetGeoData the interface that sends the sensor s geometric data to the robot function String GetConfHead the interface that sends the sensor s configuration information HEAD to the robot function String GetConfData the interface that sends the sensor s configuration data to the robot 14 2 3 Writing your own sensor Your sensor should extend from the Sensor class Yo
143. nt USARSim robot drivers are written for Pyro In summary there are three levels of control provided by the drivers The lowest level driver is robots driver utbot py It communicates with the Unreal server through a TCP IP socket The main functions in the driver are 1 Creating a connection with the Unreal server 2 Sending commands to the Unreal sever 3 Listening and parsing messages from the Unreal server 133 In the robots USARBot directory are the low level drivers __init__ py is the basic driver that provides the Pyro interface It lets the Pyro commands and data be understood by USARSim The P2AT py P2DX py PER py etc are the drivers extended from the basic driver These drivers configure the basic driver according to the individual robot For example it configures which sensor is mounted on the robot At last you will find several files in the plugins robots USARBot directory These files are the wrapper to the robot drive You can directly load these files from the Pyro GUI to add a robot into the USARSim virtual environment 13 2 3 Services To help the user to understand the data being reported by the sensors some services are added to visualize the sensor data These sensor visualizations are modified from the visualization module of PyPlayer http robotics usc edu boyoon pyplayer To load the services from the Load menu select Services Then go to plugins services USARBot directory you
144. nt is created in USARSim Replacing all the CameraTextureClient by myCameraTextureClient in the tutorial will give us a mirror effect that works online To add myCameraTextureClient go to the Actor Classes browser in Unreal Editor select myCameraTextureClient from the path Actor Info CameraTextureClient myCameraTextureClient 14 1 3 Obstacles and Victims To get a high fidelity simulation we recommend using Karma objects as the obstacles An example of adding Karma objects in a map can be found at UDN Karma Colosseum http udn epicgames com Two ExampleMapsKarmaColosseum 144 There is a known bug in UT2004 that the KActor doesn t support networks well The KNActor included in USARSim is the substitute that fixes this bug Victims are another type of objects we may need to put into the map Victims are special objects that can implement some actions The victim model built in USARSim can be loaded from the Unreal Editor To load it please open the Actor Classes browser and select the USARVictim from the following path Actor Pawn UnrealPawn xIntroPawn USAR Victim After you put it on the map you can 1 Set the mesh The default mesh is Intro_gorgefan Intro_gorgefan To change the mesh double click the victim to pop up the USARVictim Properties Then open the Display category Changing the Mesh item in this category will set the victim s mesh 2 Specify the actions In the
145. numbers in the WGS84 frame Typical output is SEN Type GPS Name GPS1 Latitude 47 40 5899 Longitude 122 18 9360 This example represents 47 N 40 5899 122 W 18 9360 The map on which GPS is used should have exactly one GPSStart object This gives the latitude and longitude of the origin of the map It also has scale factors that specify how many meters correspond to one degree of latitude or longitude at this position The scale factors may be negative to point the axes the opposite way 10 14 Robot Camera 10 14 1How the sensor works The Camera is a special sensor in USARSim The scenes viewed from the camera are captured by attaching the viewpoint to the camera in the Unreal engine USARSim currently supports any amount of cameras We can use the SET Type Camera command to control a camera s field of view and the MISPKG command to control a camera s pan tilt frame Information about the camera commands and response messages can be found in the preceding sections of this manual Every camera has a default maximum and minimum field of view If the field of view in the SET command is out of the camera s FOV range it will adjust it to the minimum or maximum FOV range We provide two ways to simulate camera feedback 1 Directly using the Unreal Client as video feedback This is the easiest way However it can t simulate the frame rate of the real robot There are two ways to directly use the Unreal Client
146. o always use stereo camera then you can 18 delete the if isStereo statement You can declare this variable config so to be able to set its value inside USARBot ini file e uuStereoSpacing is a vector that defines the eye to eye 2 spacing in unreal units 1m 250 unreal units so you can make a manual conversion Otherwise you can use the converter to automatically convert a stereo spacing parameter expressed in meters to unreal units To do so you have to add the following function in you robot class simulated function ConvertParam USARConverter converter 4 Super ConvertParam converter 5 if converter None uuStereoSpacing converter LengthVectorToUU stereoSpacing 4 else 1 uuStereoSpacing stereoSpacingi where e stereoSpacing is a vector that defines the eye to eye 2 spacing in meters You can define a vector in defaultproperties like this defaultproperties 1 stereoSpacing Y 0 05 5 cm eye to eye 2 spacing in meter 3 2 6 3 Configuration You can configure the MultiView in the USARBot ini file You will find the USARBot MultiView section that contains the following parameters e CameraNum maximum number of cameras that can be drawn at the same time e WideMode if 0 the MultiView will split the screen using a box pattern You can see in Fig 7 the pattern used for 4 cameras If gt O then the MultiView will use that number of columns before adding a raw In Fig 6 you can see
147. o the Unreal Client to set the viewpoint and then to switch back to SimpleUI After you set the viewpoint 33 and click the OK button on the message box the Unreal Client will be hidden and the control interface will appear NOTE Only press the OK button when you have set the viewpoint correctly If you pressed the OK button before you set the viewpoint you still can use the Show UT button to launch Unreal Client and set the viewpoint 5 On the control interface you can monitor the camera pictures sensor data and control the wheels and camera of the robot The usage of the control interface is a Video group In the image frame is the picture from the camera Under the frame FPS is the actual video frame rate in frames per second Width and Height is the size of the picture Show UT Hide UT button displays or hides the Unreal Client When the Unreal Client is displayed you can reset the viewpoint on Unreal Client Range Sensor Data group If your vehicle uses sonar IR range sensors or LIDAR these will be shown graphically with a color code in this box Sensor Data group The sensor data is displayed on a sensor tree You can open or close a branch to show or hide the detailed sensor data Wheels group This group controls how the robot moves The arrow buttons control the robot in the corresponding direction with a fixed speed The Faster
148. oat FrontSteer float RearSteer float Normalized bool Light bool Flip bool Where Speed float FrontSteer float RearSteer float Normalized bool Light bool Flip bool float is the spin speed for the wheels that are powered If we use normalized values the value range is 100 to 100 and corresponds to the robot s minimum and maximum spin speed respectively Otherwise the value is the absolute spin speed in radians per second float specifies the steer angle of the robot s front wheels If we use normalized values the value range is 100 to 100 and corresponds to the robot s minimum and maximum steer angle respectively Otherwise the value is the absolute steer angle in radians float specifies the steer angle of the robot s rear wheels If we use normalized values the value range is 100 to 100 and corresponds to the robot s minimum and maximum steer angle respectively Otherwise the value is the absolute steer angle in radians Indicates whether we are using normalized values or not The default value is False which means absolute values are used bool is whether turn on or turn off the headlight The possible values are True False If a robot rolls over or otherwise tips off of its wheels this will right the robot If bool is True this command will flip the robot to its wheels down position 69 Example DRIVE Speed 1
149. object is found inside the map it is used as the reference point Otherwise the ZeroZeroLocation inside the GPSSensor section of USARBbot ini is used If there is no ZeroZeroLocation inside the USARBot ini file a 83 default ZeroZeroLocation is used The next paragraphs describe how to assign a GPS coordinate to a virtual map using one of these three techniques e Reference GPS Coordinate A ReferenceGPSCoordinate object has been created so that world creators can add a GPS coordinate to a particular point in a map 1 Make sure that USRASim is compiled 2 Open the desired map in UnrealEd 3 Open the Actor Class browser and click on the Open Package icon Tf Actor Classes File View Class Textures Actor Classes Meshes Animations Static Meshes Prefabs Groups alaja ols I Us Open Package IV Placeable classes Only Actor ASCinematic_Camera AntiPortal ctor Decoration Emitter FluidSurfaceDscillator 4 Enter or Choose the file USARBot u and Click Open Open Class Package Look in E System z 0em 2 Reditorres EA SkaarjPack rcu EA Vehicles u E assaulter u EA streamlinerX u E XAdmin u My Recent S BonusPack u unrealed u Ed xEffects u Documents R Core u ig u E xGame u 3 editor EA xGame_rc u hts Engine le af E xInterface u esktop Freu EJUSARMisPkg u E Pickups 5 Gamerlay u USAR Models u A xpickups_rc u A coodKarma u dusarvictims u
150. ocity m s X 0 00 Y 0 00 2 0 00 E Camera Light On False Intensity 0 Battery 3392 Other NFO Gametype BotDeathMatch Level DM ARDA_250 TimeLimit 0 Ranges Scanner 20 0000 19 9945 20 0000 20 0000 20 0000 20 0000 20 0000 3 3438 3 3483 19 9912 20 0000 20 0000 19 99 gt Figure 16 SimpleUI 5 2 5 2 Using SimpleUI by remotely receiving pictures To run SimpleUI in remote mode we need to start image server before we launch SimpleUI Please see section 3 2 5 for information on starting the image server After the image server runs we run SimpleUI in the following steps 1 Execute SimpleUl exe which is located on ut2004 Tools SimpleUI Release directory 35 2 Similar to the step 3 in the last section we set the Command Robot and Video parameters on the interface For the Video group because we want to run SimpleUI in remote mode we need to select the Remote radio button and set the Host and Port to the image server s IP address and port number 3 Click the Start button to launch the control interface On the interface we will find the pictures are not the scenes viewed from the robot s camera This is because when we started the image server we had no robot in the world We couldn t set the robot s viewpoint at that time So we need to go back to the image server to reset the viewpoint On the ImageSrv interface we click the Show UT button to launch
151. oint This command is used to drive a joint The command looks like 72 SET Type Joint Name string Opcode string Params p p2 Where Name string string is the joint s name Opcode string string is the operation code The available codes are Angle or 0 set the extra angle in radian the joint will move Velocity or 1 set the spin speed in radian per second for the joint Torque or 2 set the torque applied on the joint Params p p2 pl is the value corresponds to the Opcode p2 only uses for the KCarWheelJoint to set the steer angle Example 1 SET Type Joint Name UpperArm Opcode Angle Params 1 25 Example 2 SET Type Gripper Name Gripper Opcode Open Params 0 8 e Control the viewports When USARSim initially starts up users can freely move around the world using the mouse and keyboard Pressing the left mouse button of the mouse attaches the view to the robot s viewport controller The robot viewport controller currently supports two configurations SingleView and QuadView The SingleView configuration provides users with a single view coming from a single camera The QuadView configuration provides user with four views giving them the possibility of viewing up to four cameras simultaneously the screen is divided up into four equal portions each of which is used for a camera When a robot is added to a world the viewport configuratio
152. olution or scan frequency too high 10 4 Odometry Sensor 10 4 1 How the sensor works The simulated odometry sensor uses the robot s left and right wheel encoders to estimate the robot s pose It describes a pose as x y position and head s theta angle relative to the start location and robot s head direction The positive x axis and y axis are the robot s head direction and right hand direction in the start location The sensor applies a very simple algorithm to calculate the pose by using the wheel s diameter left and right wheel s separation and the wheels spin angle The sensor s errors come from the diameter and wheel separation measurement error the encoder s resolution and the error introduced by the simple algorithm When we use the sensor we need to specify which wheels are the left and right wheels used in pose estimation If we don t specify the wheels the sensor will try to find and use the left most wheel and right most wheel to calculate the pose While using the sensor we can reset the sensor to use the current location and head direction as the pose estimation s reference point The command we used to reset the sensor is SET And the Opcode Params and returned response Status are listed below Table 4 Odometry sensor control command Opcode Params Status 81 RESET None OR successfully reset Failed didn t reset the sensor It may be ca
153. on of CControlDlg class in the SimpleUI source files The format of the data in the memory is defined below define FRAME_PENDING 0 define FRAME_OK 1 define FRAME_ERROR 2 typedef struct FrameData_t BYTE state BYTE sequence USHORT width USHORT height UINT size BYTE data 640 480 3 1 FrameData where state The state of the memory It can be FRAME PENDING The memory is in use by the DLL FRAME _OK The memory is ready for reading FRAME ERROR Something is wrong with the data sequence The sequence of the data The DLL only captures a new 159 width height size data picture when it gets a new sequence number You can use it to control when the DLL captures a picture The width of the captured picture The maximum width is 640 If the Unreal Client s window width is larger than 640 the DLL will not capture any pictures The height of the captured picture The maximum height is 480 If the Unreal Client s window height is larger than 480 the DLL will not capture any pictures The actual data length in the data array When the picture width is not in DWORD boundary 0 is padded to reach the DWORD boundary In this case the size isn t width height 3 The array stores the picture data The picture is stored from left to right from top to bottom A pixel is represented as Red Green Blue Each color occupies one byte 14 5 3 Using the Image Server 5 2 5 2 Its workflow is The Imag
154. ot model to help users build their own robot In the robot model every robot is constructed of e Chassis the chassis of the robot e Parts the mechanical parts such as tires linkages camera frame etc that are used to construct the robot e Joints the constraints that connect two parts together In the robot model we use Car Wheel Joint e Attached Items the auxiliary items such as sensors effecters etc attached to the robot A chassis can connect to multiple parts through joints However each part can only have one joint The attached items can be attached to either the chassis or a part The chassis or part can have multiple attached items The working flow of building a robot is to first build a geometric model for all the objects used to construct the robot Then create part wheel classes for all the robot parts wheels that extend from KDPart USARTire and a new robot class that extends from KRobot In the robot class you set the physical attributes of the robot And you also need to configure how the chassis parts wheels and auxiliary items are connected to each other Lastly if you want to add some new features not included in the robot model you will do some programming work 14 4 1 Step1 Build geometric model Essentially this step is the same as building your own arena Please refer to section 14 1 1 to learn how to build a static mesh One thing we want to emphasize 151 here is that the orientation of th
155. ow to run USARSim manually and automatically 5 2 Examples There are five controllers in the USARSim package We explain them separately in the following sections 5 2 1 The testing control interface USAR_UI is a testing interface written in Visual c 6 0 It only works on Windows You can use it to send any commands to the server and it will display all the messages that come from the server to you Follow steps 1 and 2 in section 5 1 to start the Unreal server and client and then execute usar exe This will pop up a window To use the interface 1 Click Connect button to connect to the server 2 Type the spawning robot command in the command combo box then click send to send out the command An example spawning command looks like INIT ClassName USARBot P2DX Location 4 5 1 9 1 8 where ClassName USARBot P2DX is the robot type The P2DX may be replaced by Zerg Talon P2AT P2DX ATRVJr Hummer etc The Location is the initial position of the robot Each map comes with a text file that contains recommended start points Sample startpoints for the USAR arenas are given in Table 2 3 After adding the robot to the simulator you can try to give control commands through the command combo box The messages from the server are displayed on the bottom text box To view a message double click it 4 You can also use a joystick or keyboard mouse to control the robot To do this click the Command butto
156. player 1 6 you should definitely use the latest player release currently player 2 0 3 Player is primarily used on Linux system and other POSIX platforms For Windows there is a work around to get Player 1 6 5 work with MinGW http sourceforge net mailarchive message php msg_id 15368188 For player 1 6 please follow the installation document in playerl 6 tar gz to install it For other versions of Player do NOT use the configure file in player tar gz or player1 6 tar gz You need to generate it by using GNU Autotools The installation steps are 1 Copy the usarsim directory in the player tar gz into your Player s server drivers directory So in your Player you will have a directory like server drivers usarsim 2 Open the deviceregistry cc in the server directory and add the following lines before the definition of player_interface_t interfaces ifdef INCLUDE_USARSIM void UsBot_Register DriverTable table void UsPosition_Register DriverTable table void UsPosition3d_Register DriverTable table void UsSonar_Register DriverTable table void UsLaser_Register DriverTable table void UsPtz_Register DriverTable table endif And then add the following lines into the function register_devices 26 ifdef INCLUDE_USARSIM UsBot_Register driverTable UsPosition_Register driverTable UsPosition3d_Register driverTable UsSonar_Register driverTable UsLaser_Register driverTable UsPtz_Regist
157. port 3 Viewport 3 is the Bottom Left viewport in QuadView Viewport4 string string is the name of the camera that will be attached to viewport 4 Optionally string can be set to Disable to disable viewport 4 Viewport 4 is the Bottom Right viewport in QuadView Example SET Type Viewports Config QuadView SET Type Viewports Config QuadView Viewportl Cameral Viewport2 Camera2 Viewport3 Disable Viewport4 Disable SET TypeViewports Viewportl Disable e Control a camera NOTE There is a difference between moving and controlling a camera Moving a camera i e pan tilt is achieved by controlling the mission package that the camera is attached to see the next subsection Controlling a camera is used to set its field of view SET Type Camera Name string FOV float Name string FOV float Where Type Camera Camera is a hard coded parameter Name string string is the name of the camera as described in the USARBot ini file for 74 FOV float which we want to change its field of view float is the desired camera s field of view in radians Smaller fields of view give a zoom in effect If float is zero the default field of view will be used If float is greater than the maximum field of view the maximum field of view will be used If float is smaller than the minimum field of view the minimum field of view will be used
158. pyro page PyrolInstallation to install all the packages software needed by Pyro Please remember download and install the windows version 2 After you restore Pyro you need not run make to compile it Since it uses gcc and gmake to compile files you will need these installed on your machine or the makefile will not work Furthermore it also tries to use XWindow so give up compiling it under windows Since this step only 25 affects the plugged third part robots or simulators it has no impact on USARSim After you installed Pyro you need to download and unzip pyro_win zip from the Tools section of the file release page on the USARSim repository to overwrite the files in the Pyro directory When the system asks you whether you want to overwrite the file s or not please select yes NOTE When you restore the zipped file please make sure it is restored under correct directory If your directory structure looks something like Pyro Pyro you need to move all the files under Pyro Pyro to Pyro 4 4 2 2 Linux 1 Following the Pyro Installation web page http pyrorobotics org pyro page PyrolInstallation to install Pyro 2 Download the pyro_linux tar from the Tools section of the USARSim files release page and restore it to the Pyro directory to install the USARSim plug in 4 43 Player If you are new to player and not restricted by hardware player drivers that are only available for
159. r zxvf path_to_gtki 1 9 tgz gtki 1 9 tgz cd gtki 1 9 0 configure make make install MoM M WM Mm Then you need to set the PKG_CONFIG_PATH environment variable to tell the system where GTKI is located TIP Use the following command to set PKG_CONFIG_PATH S export PKG_CONFIG_PATH path_to_gtki MOAST may now be installed Download the tar file from SourceForge or check the code out of CVS and then execute the following commands cd path_to_moast configure make NOTE If checking out from CVS you will need to issue the command boot strap before the configure command All of the code in the MOAST repository is built using autotools This allows for the configure to figure out your system s individual configuration and build the code appropriately 4 4 2 Pyro Please note that the pyro interface is out of date and no longer supported If someone would like to renew support for this interface please mail the USARSim developer s mailing list 4 4 2 1 Windows Pyro is designed for Linux Although Python the development language used by Pyro works under any OS system Pyro uses some features only supported by Linux such as Linux environment variables shell commands This makes Pyro only work on Linux We have made Pyro work under Windows The modified code can be found on pyro_win zip To install Pyro under windows 1 Follow the Pyro Installation web page http pyrorobotics org
160. r button in meter 89 10 9 RFID Sensor 10 9 1 How the sensor works The RFIDSensor simulates a RFID reader and is used to detect RFID tags that are in sensor range It also allows one to read and write their memory RFID tags can be added in the virtual world either from UnrealEd or they can be dynamically released by the RFIDReleaser effecter Refer to http en wikipedia org wiki RFID for further information on RFID technology 10 9 2 How to configure it Like any other USARSim class the sensor can be configured from the USARBot ini file In the USARBot RFIDSensor section you can set these parameters e SensingMode default Attenuation it can be Radius Obstacle Attenuation It specifies what signal propagation model to use Radius and Obstacle mode parameters e MaxRange default 3 m sensor range It s only used by Radius and Obstacle sensing modes e MaxSingleShotRange default 6 m same as MaxRange but for single shot tags Attenuation mode parameters e dBmTXPower default 36 dBm sensor TX power in dBm e dBmRXSensibility default 88 dBm sensor RX sensibility in dBm e dBObstacleAttenuation default 3 5 dB attenuation due to an obstacle Other parameters e bTraceRFIDs default false if set to true then USARSim will draw 3D lines from sensor to rfid tag slow use it only for debugging e bAlwaysReadRFIDmem default false if you want to read tag memory every time you detec
161. r case the prim shell will be started and the window where the run script was started should now be displaying a gt prompt that lets you know you are in the prim shell Other command shells may be opened by hand in additional windows Pressing a carriage return lt CR gt will print the robot s current status Entering lt CR gt will provide a list of available commands The information provided by the status message is as follows command_type The name of the executing command echo_serial_number The serial number of the executing command status The system status state Most RCS controllers run state machines This is the id of the current state line The line in the source code of the state table that is being executed source_line The location of the beginning of the current state in the source file source_file The name of the source code file heartbeat A constantly increasing number that allows you to know the system is functioning pathIndex If the system is following a path this indicates which point in the path is being servoed to tranAbs The x y z location in meters of the vehicle in absolute coordinates rpyAbs The roll pitch and yaw in radians of the vehicle in absolute coordinates tranRel The x y z location in meters of the vehicle in vehicle relative coordinates rpyRel The roll pitch and yaw in radians of the vehicle in relative coordinates 130 The available commands for th
162. r close the gripper to pick up or drop an object It currently only manipulates the KNActors the modified KActors that support network Gripper 100 uses the SET command to open or close itself The Opcode Params and returned response Status are listed below Table 6 Gripper control command Opcode Params Status Open The angle between left and OK successfully opened the gripper right fingers in radians Failed can t open the gripper It may be caused by an_ invalid command 11 1 2 How to configure it The Gripper s configuration in the USARBot ini file looks like USARBot Gripper MaxAngle 1 57 MinAngle 0 0 LeftFinger LeftFinger RightFinger RightFinger Where MaxAngle MinAngle LeftFinger RightFinger The maximum angle between left and right fingers in radians The minimum angle between left and right fingers in radians The left finger s name If we leave this parameter undefined USARSim will iterate all the parts connected to the gripper base The last found left part will be used as the left finger The right finger s name If we leave this parameter undefined USARSim will iterate all the parts connected to the gripper base The last found right part will be used as the right finger An effecter is very similar to a sensor We can mount it on the robot and use the SET command to control it An effecter is more like a dumb sensor that can t send any data o
163. r the sensor will be visually shown in the simulator Setting it to true will hide the sensor We recommend setting it to true if it s not necessary to show the sensor When you want to confirm if the sensor is placed in the correct position and has the correct direction you can temporarily set it to false The weight of the sensor in kg The relative random noise amplitude With the noise the data will be data data random noise data where random noise returns a value between noise and noise The distortion curve It is constructed by a series of points that describe the curve 95 10 12 Human motion sensor 10 12 1How the sensor works The Human motion sensor simulates a pyroelectric sensor It s simulated by finding all the victims that are in the FOV of the sensor within the testing range The closest moving victim will be checked Its distance from the robot and its motion speed and amplitude are used to calculate the probability of whether it is a human motion 10 12 2How to configure it The human motion sensor configuration in the USARBot ini file looks like USARBot HumanMotionSensor HiddenSensor True Weight 0 4 MaxRange 1000 FOV 60 Noise 0 1 OutputCurve Points In Val 0 000000 Out V al 0 000000 In V al 1000 0000 00 OutVal 1000 000000 Where HiddenSensor Boolean value is used to indicate whether the sensor will be visually shown in the simulator Setting it to true will hide the sensor
164. r turn off the headlight The possible values are True False Example DRIVE Propeller 1 0 will drive the robot forward with a propeller s spin speed of 1 rad sec DRIVE Propeller 1 0 Rudder 0 523599 will drive the robot 30 forward and to the right with a spin speed of 1 rad sec DRIVE Speed 1 0 FrontSteer 0 523599 will drive the robot 30 forward and to the left with a spin speed of 1 rad sec DRIVE Light true will turn on the robot s headlights o DRIVE AltitudeVelocity float LinearVelocity float LateralVelocity float Rotational Velocity float Normalized bool Where 70 AltitudeVelocity float is the altitude velocity i e up down If we float use normalized values the value range is 100 to 100 and corresponds to the robot s minimum and maximum altitude velocity respectively Otherwise the value is the absolute altitude velocity in meters per second LinearVelocity float is the linear velocity i e forward backward float If we use normalized values the value range is 100 to 100 and corresponds to the robot s minimum and maximum linear velocity respectively Otherwise the value is the absolute linear velocity in meters per second LateralVelocity float is the lateral velocity i e left right If we float use normalized values the value range is 100 to 100 and corresponds to the robot s minimum and maximum lateral velocity respectively Otherwise the value is t
165. real server using the appropriate arena map in the 132 Unreal world After a wait of 5 seconds to load the server it will launch the Unreal client The world files for USARSim are stored in the plugins worlds USARSim py directory NOTE here USARSim py is not a file It s a directory The file follows the INI file format A world file looks like Server Path c ut2004 App ut2004 exe LoadServer true IP 127 0 0 1 Port 3000 Map DM USAR_ yellow Location 4 5 1 9 1 8 Where Path App LoadServer IP Port Map Location 13 2 2 Robots The install path to UT2004 The application used to load Unreal Client For UT2004 it s UT2004 exe A Boolean variable indicating whether the loader needs to start the Unreal server If you already started Unreal server or you want to run the Unreal server on another machine you need to set LoadServer to false Default value is true The IP address of the Unreal server Default value is 127 0 0 1 The port number of the Gamebots Default value is 3000 The port number should be the same as the ListenPort in the BotAPLini file in the Unreal system directory more details see section 9 1 The Unreal map you want to load For yellow orange and red arenas they are DM USAR_yellow DM USAR_ orange and DM USAR_red The initial position where the robot will be spawned Please refer Table 2 or the map location files that are bundled with the maps to decide the values you wa
166. rets the player driving commands according to this steering type Interfaces Supported interfaces e position2d Required devices e simulation Configuration file options Name Type Default Meaning simulation int 1 The simulation id that specifies the USARSim robot where this device is located Odo_name String Odometry The name of the USARSim odometry sensor 13 3 2 3 us_position3d Synopsis The us_position3d driver is the same as us_position except that it uses the position3d interface Interfaces Supported interfaces e position3d Required devices e simulation Configuration file options Name Type Default Meaning simulation int 1 The simulation id that specifies the USARSim robot where this device is located 13 3 2 4 us_sonar Synopsis 139 The us_sonar driver is used to access the robot s sonar arrays Interfaces Supported interfaces e sonar Required devices e simulation Configuration file options Name Type Default Meaning simulation int 1 The simulation id that specifies the USARSim robot where this device is located sonar_name string The name of your sonar array if you have a sonar array named F1 F8 your sonar name must be F 13 3 2 5 us_laser Synopsis The us_laser driver is used to access the robot s laser sensor It only
167. robots through Player At first we explain how to plug USARSim into Player And then we explain all the Player drivers added into USARSim For additional information about Player please read the Player Wiki http playerstage sourceforge net wiki Main_Page 13 3 1 Simulation and device configuration In Player all the USARSim actuators and sensors are treated the same as the physical devices The only difference between our USARSim device driver and the physical device driver is that our driver exchanges data with the Unreal server while physical device drivers exchange data with physical devices Like all the other Player devices to use USARSim we need to define the Player configuration file In the Player configuration file we need to 1 Define the USARSim robot Before we can control a robot in USARSim we need to spawn it in the virtual world The USARSim robot definition defines how the robot is added into USARSim In detail we define where the Unreal server is which type of robot is added to the simulation and where it is spawned The complete configuration options can be found in section 13 3 2 1 136 2 Define USARSim devices In the definitions we define the parameters that help Player figure out where and how the devices are connected The definition is very similar to that of the physical devices The only exception is that instead of defining the device s connection port we define the USARSim robot where the device
168. rovides a script language Unreal Script to the game developers to develop their own games With the scripts developers can create their objects we call them actors in the game and control these actors behaviors Unreal Editor is the 3D authoring tool that comes with the Unreal engine to help developers build their own maps geometric meshes terrain etc For more information about Unreal engine please visit the Unreal Technology page http www unrealtechnology com html technology ue2 shtml 3 1 2 Gamebots The communication protocol used by Unreal engine is proprietary This makes accessing Unreal Tournament from other applications difficult Therefore Gamebots http www planetunreal com gamebots a modification to Unreal Tournament is built by researchers to bridge Unreal engine with outside applications It opens a TCP IP socket in Unreal engine and exchanges data with the outside USARSim enables Gamebots to communicate with the controllers To support our own control commands and messages some modifications are applied to Gamebots 3 1 3 Controller Controller is the user side application that is used for your research such as robotics study team cooperation study human robot interaction study etc Usually the controller works in this way It first connects with the Unreal server Then it sends command to USARSim to spawn a robot After the robot is created on the simulator the controller listens to the sensor data a
169. rt The RobotName should be the one with which the robot was registered and Port the port at which the program is listening The LISTEN word is not case sensitive Opening a connection OPEN RobotName Port HostRobot HostPort RobotName is the robot that you want to connect to and is listening for connections at Port HostRobot is the robot that wants to setup the connection and it should be listening at HostPort before sending this command to the server in order for the connection to be successfully opened if the signal strength is suitable The OPEN word is not case sensitive Sending a message SEND MessageLength Message Once a connection has been established this is used to send Message with MessageLength characters to the other endpoint of the connection This supports binary message strings The SEND word is case sensitive Closing a connection CLOSE This is used to close a connection The CLOSE word is case sensitive Getting the path loss between two robots GETSS Robot Robot2 12 This can be sent to the WSS over the control connection to get the path loss between Robotl and Robot2 The GETSS word is not case sensitive Responses All except the GETSS SEND and CLOSE command get one of two responses from the server If the command is successfully carried out the response is OK If the command could not be carried out the response is Fail ErrorMessage where ErrorMessage is a message describing why the comm
170. s Actor Classes Meshes Animations lo el Sal Ol e IV Use Actor as Parent IV Placeable classes Only 7 LiftCenter oe LiftE xit MOV PathNode ShootS pot SmallNavigationPoint Ladder PlayerStart Surface Properties 1 Selected Edit Add PlayerStart Here Add Static Mesh 2k4ChargerMeshes ChargerMeshes HealthChargermESH DS Add Path Node Here Add Player Start Here You have just placed a PlayerStart 40 f Unreal Level Editor Build UT2004_Build_ 2005 11 23 16 22 F Games UT2004 FEES File Edit View Brush Build Tools Help You can select it move it and rotate it like any other object in the editor ctrl left click drag to move ctrl right click drag to rotate Pay attention to place it not too high otherwise if you compile the map you ll receive a warning I placed this PlayerStart in the middle field 41 A Unreal Level Editor Build UT2004_Build_ 2005 11 23 16 22 F Games UT2004 PETE File Edit View Brush Build Tools Help Every PlayerStart defines a starting position and orientation But when your controller connects to USARSim how can you chose between different starting poses You can assign an identifier to every PlayerStart using the Tag field double click on the PlayerStart to open the properties window 42 tf Unreal Level Editor Build UT2004_Build_ 2005 11 23_ 16 22 F Games UT2004 Mi E3 File Edit View Brush Build
171. s Help Figure 13 Rebuilding the world 3 2 6 2 Using the MultiView As shown in Figure 14 with MultiView you can capture the camera view from many robots at the same time Figure 14 Multiple robot views using MultiView 17 Just keep left clicking in the client until you find the MultiView view You can write your own capturing image server extract the images of each robot and send them to respective clients as illustrated in Figure 15 Image Server Network USARSim i a a Controller Robot A Camera Coordinates f pos Controller Robot B Multi Rendering View Controller Robot C Controller Robot D Figure 15 Example of using MultiView with a controller Otherwise you can use the image server that comes with USARSim and split the camera images at client side MultiView can also render stereo cameras in each subview To do so you have to add this function to your robot class subclassed from KRobot Called by KRobot Tick function Stereo vision when viewing from multiview robot camera simulated function SyncMultiView 4 if ViewManager none 1 if isStereo ViewManager UpdateView ViewNum myCamera Location myCamera Rotation CameraZoom true uuStereoSpacing 4 else ViewManager UpdateView ViewNumi myCamera Location myCamera Rotation CameraZoom false where e isStereo is a boolean variable you can use to enable disable stereo rendering Obviously if you plan t
172. s used to explore AI and robotics The details of the Pyro plug in are described in section 13 2 The USARSim Player drivers are the device drivers that allow the control of robots and sensors in the simulator through Player as if they were real physical devices Player is a robot device server that gives users simple and complete control over the sensors and actuators on the robot For more information please visit Player website http playerstage sourceforge net A detailed explanation of the USARSim Player drivers can be found in section 13 3 3 2 Simulator components The core of the USARSim is the simulation of the interactive environment the robots and their sensors and effecters We introduce the three core components separately in the following sections 3 2 1 Environment simulation Environment plays a very important role in simulations It provides the context for the simulation and only with it can the simulation make sense Several specialized environments that are distributed for use with USARSim are described below Users and developers are free to create additional usage areas for the simulation 3 2 1 1 USAR Environment USARSim was originally based upon simulated disaster environments in the Urban Search and Rescue USAR domain The environments are simulations of the National Institute of Standards and Technology NIST Reference Test Facility for Autonomous Mobile Robots http www isd mel nist gov projects USAR
173. secsnadee a oe 116 12 13 1 TEP OMUCtlON tic SA Ns 116 12 13 2 CON a AEE 116 12 14 A A E AUER EI Dud h t SAR RRO 116 12 14 1 IntrOduct Onis na Bl elie he Blt ete esi BE 116 12 14 2 COMMS UTE asian da add 117 12 15 ds lan e to tank e ble 117 12 15 1 Tt dC it a 117 12 15 2 NETS UTS AEEA acu ca use dia A E 118 12 16 Kira a 118 12 16 1 IMLOU CI N it AA E Ai 118 12 16 2 A E 119 12 17 O A O E SI A O E 119 12 17 1 Introduce I ee 119 12 17 2 CONS ar ae 120 12 18 D ESE er ae Res EEE ee PEACE tna OR nT MER nt NI A 120 12 18 1 Introducido dan 120 12 18 2 A ai a dade y tics wad eee an anata deen EST 121 12 19 CTEM aK yes DA a ato ENN a A 121 12 19 1 Mtro dC do E N A E 121 12 19 2 COMPS UTS itise arete s e e r a e good a e eaS 123 12 20 ATER ODO aida tidad O Ra as 123 12 20 1 Introduction 123 12 202 Confere Ma aenea a a a A iia 124 12 21 A NI 124 12 21 1 INtPO MUCH OM daa didas 124 12 212 COMU a NAS 125 12 213 Extended USARSim command for Passarola robot 125 12 22 RuBpDOL vr ata eae eee A 126 12 23 Kendall ia A ees 127 13 14 15 16 17 18 12 23 1 Introducido EE 127 12 23 2 COTATI it Aene e a A E tues Eaa E E 128 Controller 129 E S Teac face cer ice Gage csi cte an can neta hac weet des eas oor E NT A 129 O A 132 13 2 1 Simulator and World escondo lirica einni eae 132 13 227 A a saa oea aE a A A teed 133 A ad 134 A A 135 A Lena a n i a e a a e e a leaa 135 13 3 1 Simulation and device configuration
174. section of the USARSim project page http sourceforge net project showfiles php group_id 145394 8 Extract the base files you just downloaded into the 9UT2004 folder 9 Compile USARSim by running the make bat script in the UT2004 System folder 10 From this point on you can run periodic cvs update commands to get the latest snapshot and recompile as in 8 Linux 6 Install UT2004 7 Install the patch a Download ut2004 patch at http www unrealtournament com ut2004 downloads php b Download and run the shell script http www hetepsenusret net files ut2k ut2k4 patch to install the patch For more details about usage please run i ut2k4 patch help NOTE To make the code under windows run the file make bat that is located in the system directory For linux users the file make csh should be executed NOTE In addition to the worlds the file AAA_MapBaseFiles_VX XX zip located under the BaseFiles release of the Maps package is necessary for most worlds There is a testing control interface written in C If you don t want to install any controller software you can copy USAR_UI to your machine and try it USAR_UI may be found under the Tools section of the Files release area on sourceforge USAR_UI only works on Windows You can use it to send commands to USARSim and investigate all the messages received from the Unreal server NOTE When you restore the zipped file please make sure
175. t it That means that if you sense 10 times per second you will read the memory content of all tags that are in range 10 times per second e HiddenSensor default true show hide the sensor in the simulator We recommend setting it to true if it s not necessary to show the sensor When you want to confirm if the sensor is placed in the correct position and has the correct direction you can temporarily set it to false e bWithTimeStamp default true indicates whether include the timestamp in the the sensor data message 90 e Weight default 0 4 kg The weight of the sensor in kg 10 9 3 Choosing the right SensingMode 10 9 3 1 Radius Radius mode Figure 22 only takes into account the sensor range MaxRange The sensor can read any tag in this range ignoring obstacles Traced lines if bTraceRFIDs is true all the lines are green Figure 22 Radius mode Figure 23 Obstacle mode 10 9 3 2 Obstacle Obstacle mode Figure 23 is the same as Radius but it also takes into account obstacles Tags that are in sensor range but lie behind an obstacle are invisible to the sensor Traced lines if bTraceRFIDs is true green lines for reachable tags red lines for hidden tags 10 9 3 3 Attenuation Attenuation mode Figure 25 uses a signal attenuation model It considers both distance and obstacles The signal strength is function of the transmitter power dBmTXPower and distance P dB mTXPower 2 10 5 25 2 log
176. t to represent this robot 123 a Real AirRobot b Simulated AirRobot Figure 439 AirRobot In summary an AirRobot has Four propellers One color camera that can tilt Weight 1 kg Payload 200 g In USARSim the AirRobot is equipped with One tilt only color camera The AirRobot specification is as follows 12 20 2 Dimension Length x Width x Height 0 999m x 0 999m x 0 194m Maximum altitude velocity 5 m s Maximum linear velocity 5 m s Maximum lateral velocity 5 m s Maximum rotational velocity 1 5708 rad s Configure it It s the same as P2AT 12 21 Passarola 12 21 1 Introduction Passarola the aerial blimp zeppelin robot Autonomous blimp for rescue missions http rescue isr ist utl pt from IST Instituto Superior T cnico Higher Technical Institute Portugal 124 a Real Passarola b Simulated Passarola In USARSim we use classname USARBot Passarola to represent this robot In summary Passarola has e Three propellers two for the linear movement vectoring 180 degrees 90 to 90 and one on the tail for angular movements e One color camera that can tilt e Weight 3 kg e Payload 2 Kg In USARSim the Passarola is equipped with e One tilt only color camera The Passarola specification is as follows e Dimension Length x Width x Height 4 0m x 2 0m x 2 0m e Maximum altitude velocity 2 m s e Maximum linear velocity 2 m s
177. ter class also defines some utility functions There are two lookup functions that convert the name of the opcode to the opcode and vice versa function EFFECTOR_OPCODE_TYPE GetOpcodeType string opStr function String GetOpcodeName EFFECTOR_OPCODE_TYPE eType 14 3 2 Writing your own effecter Same as section 14 2 3 Here is an example of how to implement the Animate operational code in a new effecter class NOTE You do not have to implement the Get lt Opcode Name gt Conf and the Do lt Opcode Name gt for every opcode You only need to implement the appropriate functions for the opcode you are attempting to implement 150 First overwrite the GetAnimateConf function in the new effecter class function Bool getAnimateConf out EffectorOpcodeConfig opcodeConf opcodeConf opcode EFFECTOR_OPCODE_ANIMATE_TYPE opcodeConf max Val 1 57 opcodeConf maxVal 0 return true indicates that this is a valid opcode Next overwrite the DoAnimate function in the new effecter class function Bool DoAnimate float s code for doing the opcode if code was successful return true else return false 14 4 Build your robot Usually building a robot involves a lot of programming deeply understanding Unreal network architecture and the background knowledge of mathematics and mechanics It takes a lot of time in programming and debugging To facilitate the robot building we built a general rob
178. ter the SET command has been issued The status will be OK when the camera s field of view was successfully changed Otherwise the status will be Failed Example RES Time 426 76 Type Camera Name Camera FOV 0 7853 Status OK Name Camera2 FOV 0 7853 Status OK The generic response message issued for other sensors and effecters looks like RES Time float Type string Name string Status string Where Time float float is the timestamp in the virtual world when the message is sent out It is represented in seconds Type string string is the sensor or effecter s type Name string string is the sensor or effecter s name Status string string is the status after the sensor or effecter execute a command Usually it s OK means the command is successfully executed or Failed means the execution is failed For camera s zoom in out command the status is the camera s current FOV in radians The detailed information for every sensor and effecter is listed in sections 10 and 0 Example RES Time 61 20 Type Odometry Name Odo1 Status OK 9 4 Commands In USARSim all the values in the commands are case insensitive However the data_type and names are case sensitive and the format must be exactly followed The supported commands are e Add a robot to UT world INIT ClassName robot_class Name robot_name Location x y z Rotation r p y W
179. that is runing USARSim Go to the place where you store the configuration file execute the following command player lt config file gt Start your Player client For example you can run the Player visualization tool plyerv In Linux KDE UT2004 may not support switching focus to other applications As a solution we launch UT2004 on another console to let user switch between UT2004 and other applications 32 Note Player uses absolute camera control By default all the robots in USARSim use relative camera control You need to change the USARBot ini file to let Player work well For details of changing camera control mode please read section 12 5 2 5 SimpleUI SimpleUI is an example user interface under Windows It may be downloaded from the tools section of the USARSim file repository and should be placed in the UT2004 Tools directory To successfully use it please make sure the FreeImage dll Info html and SimpleULexe are in the same directory Also the file hook dll MUST be in ut2004 exe s directory Besides directly using Unreal client as the video feedback this interface demonstrates how to use the video pictures on the user interface SimpleUI can obtain video pictures either through locally capturing the Unreal Client or receiving them from the image server The details about getting and using video pictures are explained in section 10 13 and 14 5 In SimpleUI we simply display the video pictures without
180. the command was successfully implemented or not 149 function Bool DoActivate float val function Bool DoAnimate float val function Bool DoFire float val function Bool DoRelease float val function Bool DoReset float val function Bool DoNOP float val The virtual Get lt Opcode Name gt Conf functions defined in Effecter uc This function enables the base class to retrieve configuration information from a referenced argument EffecterOpcodeConfig structure It returns a Boolean to indicate whether the opcode is implemented or not function Bool GetActivateConf out EffectorOpcodeConfig opcodeConf function Bool GetAnimateConf out EffectorOpcodeConfig opcodeConf function Bool GetFireConf out EffectorOpcodeConfig opcodeConf function Bool GetReleaseConf out EffectorOpcodeConfig opcodeConf function Bool GetResetConf out EffectorOpcodeConfig opcodeConf function Bool GetNOPConf out EffectorOpcodeConfig opcodeConf The Effector Opcode Configuration Structure is a structure is used by the FormatOpcodeConf to produce the appropriate CONF message for a given opcode Therefore this structure should be used by derived classes to hold configuration information and will enable derived classes to maintain appropriate information for each opcode function String FormatOpcodeConf EffectorOpcodeConfig opcodeConf struct EffectorOpcodeConfig var EFFECTOR_OPCODE_TYPE opcode var float max Val var float minVal E The effec
181. the Unreal Client Go to the Unreal Client we attach the viewpoint to the robot we just spawned in the world Then we click the Hide UT button on ImageSrv interface to hide the Unreal Client When we switch to the SimpleUI interface we will get the correct camera pictures 4 Follow the usage introduced in the previous section to control the robot and its camera 5 3 Getting Starting Poses From Maps 5 3 1 Introduction A common question when using USARSim is where I spawn the robots Usually you can find starting positions using the editor or directly in the UT client with the showdebug console command This approach isn t difficult but cannot be automated You have always to provide starting positions manually to your robot controller If you change the map you will need to recompile the controller or to change some initialization file That s why we introduced in our lab the following USARSim command GETSTARTPOSES After connecting to USARSim you can send this command and you ll receive a string with all available starting positions for that map This way starting positions are a map property and the controller can chose automatically where to start the robots 5 3 2 Adding starting poses to the map You have to populate the map with starting poses Any NavigationPoint will do however it s much easier to use PlayerStarts because you can see and adjust the orientation in the editor 36 4f Unreal Level Editor Build UT2
182. the sensor name Prob float float is the probability of it s human motion Example SEN Type HumanMotion Name Motion Prob 0 81 o Sound Sensor SEN Type Sound Name string Loudness float Duration float Where Name string string is the sensor name 57 Loudness float float is the loudness of the sound Duration float float is the duration of the sound Example SEN Type Sound Name Sound Loudness 17 22 Duration 6 63 e Geometry Information For a ground vehicle the geometry information message looks like GEO Type GroundVehicle Name string Dimensions x y z COG x y z WheelRadius float WheelSeparation float WheelBase float Where Type GroundVehicle The Type parameter is hard coded and will always be GroundVehicle for a ground vehicle GEO message Name string string is the robot s name Dimensions x y z x defines the robot s length y defines the robot s width and z describes the robot s height Please note that these values are in meters COG x y z x y and z identify the position of the center of gravity in meters calculated from the chassis origin WheelRadius float float is the radius of the robot s wheels in meters WheelSeparation float float is the wheel separation in meters The wheel separation defines the distance between two wheels along the length x axis o
183. this is the pointer of your application 158 14 5 2 Capturing Unreal Client In USARSim Hook dll provides help to get the scenes from the Unreal Client This DLL uses Detours technology http www research microsoft com sn detours to capture the back buffer of DirectX 8 x and store it as a raw picture in a block of shared memory To use this DLL we need to 1 Attach the DLL to the Unreal Client We can use the detours function DetourCreateProcessWithDILO to combine Hook dll to the Unreal Client For more details about this function please read the withdll example that comes with Detours You can also find the example code in the SimpleUI source files The LoadUTO function of CControlDlg class is the exact function that attaches Hook dll to the Unreal Client and then launches it Because the Hook dll needs to catch the Direct3DCreate8 function the DLL must be attached to the Unreal Client before the Unreal Client runs This is the reason why we use the Detours function DetourCreateProcessWithDIIO 2 Get the address of the shared memory getFrameData is the function provided by Hook dll that tells you the address of the shared memory To get the memory address we need to i Get the module handle of Hook dll by using LoadLibrary ii Get the function address of getFrameData by using GetProcAddress iii Call the function getFrameData to get the memory address The example code can be found at GetpfFrameData functi
184. tion settings one can define the accuracy of the track simulation of the Tarantula model The tracks of the Tarantula model are simulated by a series of tires which is computationally complex on the Unreal engine When the number of tires is large the simulation might become unstable due to the computational complexity Hence the model enables one to adjust the number of tires utilized for the simulation of one track according to your specific needs Note that with increasing number of tires the quality of the simulation increases however the computational complexity increases as well The adjustment can be done in the USARBot ini file in the sections USARBot TarantulaFrontTrack and USARBot TarantulaRearTrack respectively 12 11 Zerg 12 11 1 Introduction The Zerg robot is a 4WD 4 wheeled robot which has been developed and deployed by the team Rescue Robots Freiburg during RobotCup05 The simulation model was developed at University of Freiburg and has been further improved and merged into USARSim by the University of Pittsburgh The character of this model is simple In USARSim we use classname USARBot Zerg to represent this robot a Real Zerg b Simulated Zerg Figure 35 Zerg model In summary the Zerg has e Four wheels e Differential steering In our simulation it is equipped with e PTZ camera The Zerg specification is as follows e Dimension Length x Width x Height 31 12cm x 41 54cm x 12 1l1cm e Whe
185. tly connect to the appropriate NML buffer please see the NML tutorial located at http www isd mel nist gov projects rcslib the second is to utilize one of the provided shells and the third is through the RCS Diagnostics tool An example of directly connecting to NML buffers is provided by the nmlPrint program located in the moastBaseDir devel src tools directory This program prints out the content of a selected buffer to the screen By piping its output to other programs such as gnuplot sensor displays and graphs are possible Figure shows the result of running the command spPlot gnuplot where spplot is a shell script located in MOAST s bin directory that runs nmlPrint on the buffer servoSPLinescanl This buffer contains the data received from the Sick LMS sensor on robot 1 and like all NML buffers is available to any computer that has a network connection to the system running the MOAST USARSim middle ware The actual location of the buffer is invisible to the application that is connecting to the buffer 129 Figure 40 Image from the unreal Client and resulting graphical display of Sick LMS readings provided by nmlPrint and gnuplot The building and trees are clearly visible in the plot The second technique is to utilize the command shell that is started by the run script A command shell exists for each level of the hierarchy and the run script automatically starts the highest level command shell that is appropriate In ou
186. u may add your own sensor parameters and these parameters should be in the API interface s units and coordinates In the function ConvertParam you transfer them to Unreal s units and coordinates Then you may override the Getxxxx methods to return your own data in a string When you generate this data string you need to convert the units and coordinates back to the API interface s units and coordinates 148 The robot model will call the isType and isName functions to find out whether the sensor is the current sensor it needs to process By default these two functions do simple string matching You can override them to do some advance things like considering super type and sub type Although there are Noise and OutputCurve parameters in Sensor class it does nothing about the noise data simulation and data distortion simulation It s your responsibility to simulate them in the GetData method NOTE You MUST use the converter object to do all the unit and coordinate conversion for flexibility and consistency reasons Always use isType and isName function in your code to do the type and name matching 14 3 Build your effecter This section of the manual will provide you an overview of how to implement a new effecter in USARSim In the first subsection an overview of the functions and structures in the Effecter base class will be presented The second subsection will show how to implement on opcode in a new effecter class
187. ure 21 Singularity Representation 10 5 2 How to Configure it The common bHiddenSensor and bWithTimeStamp variables are included in the USARBot GPSSensor section of the USARBot ini file that when set to true hide the sensor and add time information to the GPS message respectively In addition the amount of noise injected into the sensor is adjusted with the maxNoise and minNoise variables The maxNoise variable dictates the maximum amount of noise injected in the sensor when the receiver can only use four satellites The minNoise variable dictates the maximum amount of noise injected in the sensor when the receiver sees the maximum number of satellites twelve Please note that the outdoor worlds have to follow the typical skybox procedure of using the wm_texture sky texture all around the world this texture will then be replaced by the world s skybox Failure to comply with this standard will result in a non working GPS sensor The main problem with using a GPS sensor in a virtual environment is the mapping of a virtual location to a real one Since USARSim worlds do not inherently have a GPS coordinate associated with them the GPS sensor provides three ways to allocate a GPS coordinate to a virtual world 1 adding a special object in the virtual world a ReferenceGPSCoordinate object 2 using the USARBot ini file or 3 modifying the actual GPSSensor class A GPS reference point is acquired in this order if a ReferenceGPSCoordinate
188. used by an invalid command 10 4 2 How to configure it We configure the odometry sensor in the USARBot ini file The configuration looks like USARBot OdometrySensor HiddenSensor true bWithTimeStamp Fal Weight 0 4 ScanInterval 0 2 se EncoderResolution 0 01 LeftTire LeftFWheel RightTire RightFWheel Where HiddenSensor bWithTimeStamp Weight ScanInterval EncoderResolution LeftTire RightTire 10 5 GPS Sensor Boolean value is used to indicate whether the sensor will be visually shown in the simulator Setting it to true will hide the sensor We recommend setting it to true if it s not necessary to show the sensor When you want to confirm if the sensor is placed in the correct position and has the correct direction you can temporarily set it to false Indicates whether the timestamp in the Unreal Engine will be associated with the sensor data The weight of the sensor in kg The time interval in seconds between pose estimates The minimum wheel spin angle the sensor can recognize in radians The left wheel that will be used in pose estimating If it is a null string the left most wheel will be used The right wheel that will be used in pose estimating If it is a null string the right most wheel will be used 10 5 1 How the sensor works The GPS sensor finds the current robot position in meters and converts it to latitude and longitude The GPS sensor follows a modified version of the NMEA 183
189. ut We introduce all the effecters in this section 11 2 RFID Releaser 11 2 1 How the effecter works The RFID releaser drops a RFID tag in the virtual world The tag s ID is automatically assigned by the releaser in the range of 0 1000 to make sure that all of the IDs are unique The IDs greater than 1000 are reserved by the system You can use these reserved IDs for special purposes preplaced in world maps The releaser uses the SET command to drop a RFID tag The Opcode Params and returned response Status are listed below 101 Table 7 RFID releaser control command Opcode Params Status Release None OK successfully dropped a tag Failed can t drop tags It may be caused by an invalid command or lack of tags coe int number of tags left to dispense TagsRemaining None 11 2 2 How to configure it The RFID releaser s configuration in the USARBot ini file looks like USARBot RFIDReleaser Weight 0 4 NumTags 10 Where NumTags The number of tags that the releaser carries Weight The weight of the effecter in kg 11 3 Roller Table 11 3 1 How the effecter works A roller table is an effecter that is constructed of a roller table base with a series of controllable roller mounted on the surface of the table This effecter enables items to be moved back and forth on the table and is commonly used in manufacturing applications to load and unload packages on and off the tab
190. veral link segments A mission package state message looks like MISSTA Time float Name string Link int Value float Torque float Link int Value float Torque float Where Time float Name string Link int Value float float is the UT time in seconds It starts from the time the UT server starts execution String is the name of the mission package This parameter gives the link number that will be described by the next two parameters Value and Torque Please note that you will have as many as these parameters as you have links This parameter has two possible meanings If the link being described is a prismatic joint float gives the distance in meters 32 from the original position of the link If the link being described is a revolute joint Torque float The Torque parameter gives the current torque of the link being described Example MISSTA Time 102 09 Name CameraPanTilt Link 1 Value 0 0000 Torque 20 0000 Link 2 Value 0 1600 Torque 20 0000 e Sensor message Every sensor message starts with SEN After it is an optional Time segment Time float that reports the current time in seconds in the virtual world Whether the Time segment will appear or not is decided by the sensor s bWithTimeStamp variable For details information please read section 10 o Range Sensor SEN Type string Name string Range float Name string Range float
191. very pair defines a part and how it connects to its parent The order of the pairs is always from the root usually it s the robot platform to leaf terminal In section 12 we will give detailed instructions to control the different parts of a mission package A mission package s state is described as a series of the part s absolute angle from its initial mount position We use the MISSTA message to deliver this information This message is explained further in the next section 8 Effecters Effecters are defined in the USARSim as robotic subsystems that alter the state of the environment through direct interaction with objects and features found in the environment These subsystems differ from the mobility subsystem and the mission package subsystem in that they usually don t require joint level control and are used to perform specific tasks like grab or release object in the environment Because these devices are task specific it is hard to formulate generalized control and status messages for these systems In order to enable the possibility for autonomous control of these subsystems USARSim has imposed a restricted vocabulary that defines opcodes that are to be used by all effecters This enables new effecters to develop as along as they honor the interface by implementing one or more of the opcodes Opcodes in USARSim are enumerated types that define different actions that an effecter can perform The current opcodes that are defin
192. x UT2004 StaticMeshes I UB_meshes usx UT2004 StaticMeshes spqrMapsMeshes usx UT2004 StaticMeshes spqrRobotMeshes usx UT2004 StaticMeshes TalonMeshes usx UT2004 StaticMeshes USAR_Robots usx UT2004 StaticMeshes USARSim_Hummer usx UT2004 StaticMeshes USARSim_Sedan usx UT2004 StaticMeshes USARSim_Submarine usx UT2004 StaticMeshes Sub usx UT2004 StaticMeshes USARSim_Cars usx UT2004 StaticMeshes Hummer usx UT2004 StaticMeshes USARSim_Vehicles_Meshes usx UT2004 StaticMeshes USARSim_VehicleParts_Meshes usx UT2004 StaticMeshes USARSim_Objects_Meshes usx UT2004 StaticMeshes USARSim_LeggedRobots_Meshes usx UT2004 System Hook dll UT2004 System make bat 21 UT2004 System make csh UT2004 System usar_c bat UT2004 System usar_s bat UT2004 System usar_t bat UT2004 System USARBot ini UT2004 System USARBotAPL ini UT2004 System USARSim ini UT2004 System BotAPL ini UT2004 Textures UB_textures utx UT2004 Textures SEERS _girls utx UT2004 Textures SEERS_gWounded utx UT2004 Textures SEERS_Wounded utx UT2004 Textures SEERS Textures utx UT2004 Textures SEERSTexturesSG utx UT2004 Textures spqrMapsTextures utx UT2004 Textures spqrRobotTextures utx UT2004 Textures TalonTexture utx UT2004 Textures USAR utx UT2004 Textures submarine utx UT2004 Textures USARSim_Cars utx UT2004 Textures Hummer utx UT2004 Textures USARSim_Vehicles_Textures utx UT2004 Textures USARSim_VehicleParts_Textur
193. za Italy The first version of RFID tag and sensor from Alexander Kleiner at University of Freiburg Germany The current version RFID tag and sensors from Mentar Mahmudi at International University Bremen Germany Version 1 of the Victim sensor and tags from Stephen Balakirsky at National Institute of Standards of Technology USA Version 2 of the Victim sensor by Ben Balaguer GPS sensor from Tyler Folsom University of British Columbia Canada Modified GPS from Ben Balaguer University of California Merced 162 Radio Model e Yashodhan Nevatia from International University Bremen Germany Tools e ImageServer from Jijun Wang at University of Pittsburgh USA Modifications to allow client only operation by Marco Zaratti e MultiView from Marco Zaratti e Omni view camera from T Schmits of the University of Amsterdam 18 Acknowledgements This simulator was developed under grant NSF ITR 0205526 Katia Sycara and Illah Nourkbaksh of Carnegie Mellon University and Michael Lewis of the University of Pittsburgh Co PIs Elena Messina and Brian Weiss of NIST provided extensive assistance Joe Manojlovich Jeff Gennari and Sona Narayanan contributed to the development of the simulator Eric Garcia and Stephen Balakirsky edited this document 163

USARSim Manual

Contents

Download Pdf Manuals

Related Search

Related Contents