P8 Manuals - Mathieu Venot
Contents
1. Vues 3D 3D views Vue d un cadre type Adaptative Aggregation Based Building System An alternative to large scale 3D printing WIREBOT ARDUINO CODE Charles BOUYSSOU Oswald PFEIFFER Mathieu VENOT 24 C WireBot Arduino Code Manuel d Utilisateur User Manual R glages Settings Libraries AccelStepper controle moteurs Servo controle pince VirtualWire optionel controle sans fil Libraries AccelStepper stepper control Servo gripper control VirtualWire optional wireless control Pins 2 Step 1 Equipement Equipment ou Arduino DUE VS Arduino DUE Direction 6 Direction M2 7 Direction M3 Pins 2 Step MI 3 Step M2 4 Step M3 5 Direction M1 6 Direction M2 7 Direction M3 Arduino IDE 1 6 4 Arduino IDE 1 6 4 Variables BUFFSIZE 256 Serial begin 9600 vitesse baudrate Variables BUFFSIZE 256 Serial begin 9600 baudrate speed 25 C WireBot Arduino Code Manuel d Utilisateur User Manual Ligne de commande Command line ex 100 200 300 0 a b vitesse moteur 1 decimal pas s b vitesse moteur 2 decimal pas s c vitesse moteur 3 decimal pas s i type d instruction entier O 1 2 1 O mise en route des moteurs 1 ouverture de la pince 2 fermeture de la pince 1 arret des moteurs a speed motor 1 float step s b speed motor 2 float step s c sp2. NS 17 4 SATA L SISK LIZA D SA IK IV D 2 AISNE ISLE AS LISEZ eA AZA PLAN AN D NZ RS PASS ENI Wa aa P Ses OTETHIA SRE PRISES SESS ZI ar CA R SITL LTA STA 143 m M thode des forces concourantes statique graphique Method of concurrent forces static graphic X_ Chapter Proposition pour le controle Control proposal Anchor point Cam 1 O Charles BOUYSSOU Oswald PFEIFFER Mathieu VENOT 20 Anchor point 2 Proposition pour le controle Control proposal Variables Vitesse d enroulement maximum permise au robot Vmax Relation d asservisement propos e Slaving relationship LT 12 AE 12 VAx Ax Vmax if dist Pos Pos P OX Vmax O Principe L acquisition rapide des cam ras 3D 25 FPS permet de positionner pr cis ment le robot dans l espace de mani re relative son environnement par l application continue de la boucle d asservissement Principle The rapid acquisition of 3D cameras 25 FPS can precisely set the robot in the space relativly to its environment by the continuous application of the control loop feedback 21 Vues 3D 3D views Vue d un cadre type Frame type view Charles BOUYSSOU Oswald PFEIFFER Mathieu VENOT Z gt D gt _ E C LL
3. assemble coalesce grow or generate architectures by themselves 2 It is obvious that the development of computer simulations has been a major factor in that shift allowing the interaction of a large amount of discrete elements often referred to as agents Simulations based on this principle are commonly known as Multi agent Simulations in which because we only describe the rules that guide individuals we can only roughly predict the global behaviour of a swarm thus the exact position of an agent at a given moment in time t remains unpredictable The emergent complexity of such a stochastic approach generally offers a positive and a negative side for architects and designers a high degree of redundancy and self organization on one hand but a problematic fabrication process due to the abstract geometry of the resulting forms on the other For these reasons working directly on concepts such as redundancy self organization and as will be discussed here aggregation within the fabrication process seems a more logical approach than designing using multi agent systems and only posthumously trying to recreate the form as accurate as possible This last approach Is in fact highly contradictory and raises a lot of questions such as When should the simulation be stopped Do we have to build the resultant form exactly as is Is there a disparity between the simulated materials behaviour and the real one s Etc Recent research on large scale 3
4. being developed for dropping the aggregates offers several important advantages compared to a regular three axis CNC For this particular case the research has been focused on portability modularity and compactness All the mechanical parts are grouped inside the robot s head that way setting up the system in any environment is incredibly easy you Just need to attach the three cables to three points in space these three points form a triangle that defines the system s working space The modularity also comes from the fact that there s some space remaining on the robot s head allowing the user to add three more cables to improve the robot s precision and stability if needed The fact that we can arbitrarily define the anchor points and can ignore cable overlay a problem that plagues cable robots where the cable winds onto Itself causing massive imprecision by effectively changing the spindle s radius is afforded by a servitude algorithm that allows us to avoid absolute coordinates What this means is that the robots operations are controlled by a feedback loop that determines the difference between the desired action and the actual one This workflow uses machine vision and Is very close to the construction correction routine that was briefly outlined earlier the only difference here resides in the fact that the vision Is not based on the same algorithm Here the space and the position of the robot s head are acquired using markers specific p
5. floor 2 One or several aggregates can fall or the target spacing generated by the initial shape discretization can be too wide for the chosen aggregate density resulting in a lack of density at some points of the structure compared to the ideal simulation 3 The target spacing generated by the initial discretization can be too dense for the chosen aggregate density resulting in emergent artefacts due to a too high amount of modules dropped at the same place The computer vision uses two Kinect cameras to scan the environment in real time this generates coloured points in space resulting from the interpolation of the camera colors and of the distances given by the infrared ray casting Every n aggregates dropped the software will isolate the scanned aggregates using their colour the best way to do that is to use the equivalent of a chroma key before comparing their actual positions to their theoretical positions to solve errors 2 and 3 Obviously it is nigh on impossible to identify single elements from a collection of aggregates so in order to determine if a target has been missed error type 2 the software checks its distance to the closest scanned point of the ensemble before filling the structure s holes We use the same process for the error type 3 but this time the software is also checking the heights of the scanned points to know If some elements have been dropped above the highest targets reached so far if AABBS Th
6. instructions Because of that a software able to deal with real time environment data acquired from computer vision needs to stay connected to the robot As it also deals with aggregates of different materials sizes and densities that are potentially customizable it was chosen to encode a specific voxelizer software on Processing able to first of all guide any user through the different discretization steps material assignment and construction settings then run the robot in real time while returning information and statistics about the construction process to the user This also allows him to see the disparity between the theoretical model the system s production target and the current state of the construction process Different voxelization strategies are established for example it can be based on a recursion using three sizes of modules that way the centre of the shape can be filled with large aggregates while the borders can be built with smaller modules giving a higher resolution finish Another possibility is to use bigger aggregates at the bottom of the construction strengthening its base and smaller aggregates at the top or for crossings ADAPTATIVE COMPUTER VISION At the heart of the AABS resides the construction process s artificial intelligence So far it is able to detect and deal with three different error types 1 An aggregate can fall miss or rebound off its target instead of hooking on and stay isolated on the
7. while maximizing the cohesion power of the aggregates through their geometry The method used to find the best possible geometry of the aggregates was based on a fitness analysis through a high number of physical simulations Three families of wireframe parametric geometries star like forms pyramid like shapes and branch like structures each of which contained six different variations were tested in a proprietary benchmark developed in Unity3D to define a fitness value This benchmark ran in two sequential iterations ten generations of ten modules dropping for a first then a hundred generations of fifty modules for the second iteration Once the two best module options were identified a second phase consisted of designing three different modules for three different types of density in a more detailed way based on these results which revealed the potency of using a hook and loop typology just like hook and loop fastener in order to maximize the cohesion A new physical simulation set was then conducted to determine the best possible size of hooks and loops The aim of this second phase was also to make the shapes easy to produce with a laser cutter In the end each aggregate was made of three planar slotted cardboard pieces and the cutting and assembling process took about two minutes per module for a bounding box of 15x15x15 centimetres and minimal material COSIS This fast and cheap solution served well for testing purposes but can t be
8. 060 html Alimentation Variable 24V 24V Variable Power Supply http www selectronic fr alimentation variable a decoupage 0 a 30v 0 a 20a html Profil s ELCOM ELCOM Structure Frame M canique HPC HPC Mechanics R glages Settings Moteurs NEMA 2400 N cm Pair de cables 1 Vert amp Rouge Pair de cables 2 Bleu amp Jaune NEMA Steppers 2400 N cm Cables pair 1 Green amp Red Cables pair 2 Blue amp Yellow Controleurs Moteurs 6A Microstep 1 2 Courant de sortie 3 1A Stepper Drivers 6A Microstepping 1 2 Current Output 3 71A Alimentation Variable 24V Voltage de sortie 24V Courant de sortie 6A 24V Variable Power Supply Voltage Output 24V Current Output 6A Cable acier 2mm Steel cable 2mm D multiplication 2 1 Ratio 2 1 B WireBot Cable Robot Manuel d Utilisateur User Manual Sch mas l ctriques Electric schematics Stepper 1 ROB 08420 A C St Q EasyDriver_ v44 Power plug n Arduino 1 Stepper 2 ROB 08420 OUTIA OUTIB OUT2B Arduino OUT2A no Rev3 D9 PWM D10 PWM SS D11 PWM MOSI D12 MISO D13 SCK Stepper 3 Transceiver SLEEP OUT2B RESET EasyDriver v44 Contr le des moteurs Arduino Controleurs Moteurs Moteurs pas pas Motor control Arduino Stepper Drivers Stepper Motors Servo Receiver Arduino no Rev3 D9 PWM D10 PWM SS D11 PWM MOSI D12 MISO D13 SCK Contr l
9. An alternative to large scale 3D printing MANUEL MANUAL i Charles BOUYSSOU Oswald PFEIFFER Mathieu VENOT AABBS Theory Explanation Article Th orique Theorical Paper The Newtonian paradigm places the emphasis on external forces gravity natural selection the market and so on Taking nonlinearity into effect means we concentrate more on the system in evolution the developmental system of the organism in economics the nature of society and the people who make it up It does not as do relativity and quantum mechanics introduce entirely new scientific principles but it can completely alter the direction of our research all the same Peter Saunders 1 DEALING WITH THE UNPREDICTABLE Dealing with unpredictable materials and emergent systems has been at the heart of an important number of Architectural research projects over the past ten years The study of natural systems and how they could influence the way we design our own human constructions brought a new paradigm to architecturaldesign we do not have to design everything in a top down approach we can choose instead to declare a set of rules for a system and then let the system itself determine its evolution Ren Doursat Hiroki Sayama and Olivier Michel theorize the role of these potential meta designers in A review of morphogenetic engineering meta designers will focus on creating local mechanisms that allow small agents or components to
10. D printing 3 leads us to believe that additive manufacturing will be an important breakthrough for architecture This process shows a lot of advantages from the removal of the need for formwork to an increased freedom in the shapes definition which together give rise to the possibility of optimizing the topology of the elements produced and eventually to an autonomy of the construction process Charles BOUYSSOU Oswald PFEIFFER Mathieu VENOT 2 But it also comes with a few inherent disadvantages given that the printing process in itself is slow elements needs to be factory made with heavy equipment the machines building envelopes are limited and it is difficult to make it bigger once it has been built The system that will be presented here referred to as Autonomous Aggregation based Building System AABS does not claim to overtake large scale 3D printing it should simply be seen as an alternative to this type of construction process and strives to prove that additive manufacturing doesnt necessarily have to mimic the increasingly familiar desktop 3D printing workflow or even use continuous material like melted plastic concrete or clay Practically the AABS consists of building a given shape by dropping singular aggregate elements with a cable robot As we know the unpredictable nature of aggregates can induce errors in the creation process an element could down it could bounce off its target instead of hooking on etc Therefor
11. TI ff SETTINGS TRAHSLATE _KIMECTI SCALE_OOWH SCALE UP ROTATE _KIHECTI POIHTS_CORRECTIOH EHVIROMMEHT_SHOUI TRAHSLATEZ _KIHECTI R TRTEZ _ TRAMSLATEH_KIHECTZ SIMULATION CONSTRUCTIOH ROTATER_ KIMECTE SIMULATIOH CONSTRUCTIOH TRAMSLATE Y KIMECTA 100 00 DROP_OFF SET ROTATE _KIHECTZ TRAMSLATEZ_KIHECTZ FOTATEZ_KIHECTZ EXPORTS PHG_ERPORT RESET CODE_ERPORT COHTROLS_KEYS_HELP ff CONSOLE NU Kinect Only one kinect 15 connected You can calibrate one kinect and track a color but you will not be able to run any physical test One kinect on two 15 ready User selected C Users Oswald Google Orive PB P10 DK 2015 Slicer Oswald 1505041 Synthesis impo rtMesh walltest stl Mesh aligned translation x 1959 6 y Br 191 z Copyright 2015 Il Copyright 2015 Copyright 2015 An alternative to large scale 3D printing WIREBOT CABLE ROBOT Bor Charles BOUYSSOU Oswald PFEIFFER Mathieu VENOT WireBot Cable Robot Manuel d Utilisateur User Manual Equipement Equipment Moteurs NEMA 2400 N cm NEMA Steppers 2400 N cm http www selectronic fr c coffrets dissipateurs quincaillerie moteurs pas a pas bipolaires moteur pas a pas modele pap 20 html Controleurs Moteurs 6A Stepper Drivers 6A http www selectronic fr c robotique moteurs accessoires control modules commandes de moteurs pas a pas module de commande pour moteur pas a pas cw8
12. atterns that looks a bit like QR codes In order to pick up the aggregates the robot s pendulous head uses a simple claw at its bottom that is ample thanks to the aggregates own hooks It will then drop them one by one from the bottom to the top The electronics parts are controlled by an open source Arduino PCB so instructions for the motors can be sent using Arduino software or Processing which as we will describe in the next section has its own advantages SOFTWARE GENERAL USER INTERFACE In 3D printing the discretization of a shape and its conversion to instructions for the printer is provided by a software called a slicer As the name suggests It slices a 3D model typically a stl file into a G code like file The user then has two options to execute this code either transferring this file to the printer via an external storage device or connecting a computer directly to the printer to read the file in real time Recently a new generation of slicers have appeared called voxelizers because the discretization of the 3D object file works with voxels 3D pixels instead of slices It is an interesting approach because the printer deals with a three dimensional array of small blocks of material which makes multi material model fabrication much more viable The AABS needs a constant feedback because it s a dynamic system that can t Charles BOUYSSOU Oswald PFEIFFER Mathieu VENOT 6 work with a simple series of linear
13. considered a long term material to build architectural objects in itself There are in fact a very large amount of other possibilities for this some of which have already been tested like using the aggregation as a substructure for a lycra epoxy shell that can even be reinforced with glass Charles BOUYSSOU Oswald PFEIFFER Mathieu VENOT 4 fibres while others need more financial Investment and research into their application but possess a real construction potential such as metal aggregates as a support for shotcrete Although these two options are based on mechanical aggregation chemical aggregation could be another interesting possibility because it would not require any subsequent solidification 4 See the two following papers DIERICHS Karola MENGES Achim Aggregate structures Material and machine computation of designed granular substances Institute for Computational Design University of Stuttgart John Wiley amp Sons 2012 DIERICHS Karola MENGES Achim Material Computation in Architectural Aggregate Systems Institute for Computational Design University of Stuttgart ACADIA 2010 5 This project can be seen on the Minimaforms website http minimaforms com imogenheap 6 This project can be seen on the Gramazio amp Kohler Research website http gramaziokohler arch ethz ch web e lehre 276 html AABBS Theory Explanation Article Th orique Theorical Paper HARDWARE ROBOTICS The cable robot that is
14. dback and thus allowing the emergence and self organization of the growing structure Charles BOUYSSOU Oswald PFEIFFER Mathieu VENOT 8 An alternative to large scale 3D printing AABBS SLIC3R TOOLPATH SOFTWARE OFF THE HO Autonomous Ageregi VISUALIZATIOH LUIREFRAME f MORELIZATIOH 10 00 FILL_FOINTS f SETTINGS SCALE_DOLUH P lHTS CORRECTION if SIMULATION SIMULATION f EXPORTS PHG_EXPORT CODE _EHP RT s 20 588234 ee 10 e launching simulation Hide mesh CPS Simulation LOG created CPS Main Solid mesh display stance 957 63 meters Wireframe mesh display Showing points he robot speed is 1 0m s it would take h 57s 15m nn dules 189 CPS Simulation dules outside the mesh 82 Charles BOUYSSOU Oswald PFEIFFER Mathieu VENOT A AABBS Slic3r Toolpath Software Manuel d Utilisateur User Manual Equipement Equipment Processing 3 0a5 Processing 3 0a5 Java 1 6 Java 1 6 Manuel Manual 1 Import d un mesh a l ouverture Mesh Importation at the software opening 2 R glage de l espacement des cibles Setting targets spacing 3 Acquisition des cibles Targets acquisition A AABBS Slic3r Toolpath Software Manuel d Utilisateur User Manual 4 Correction ventuelle des cibles mal positionn es 7 Calibration des kinects Possible correction of the bad targets Kinects calibra
15. e de la pince Arduino Servo Recepteur optionel Gripper control Arduino Servo Optional Receiver Charles BOUYSSOU Oswald PFEIFFER Mathieu VENOT 16 Montage de la structure Structure assembly LT P Modular SLT Fower and control box Winding system Motor Vue clat e Exploded view B WireBot Cable Robot Manuel d Utilisateur User Manual Espace de travail Workspace m v s Unreachable area Reachable area A a S a a a BU a TA a a Pai wr AE A 4 n P i 9 Espace utile atteignable du WireBot WireBot reachable area Charles BOUYSSOU Oswald PFEIFFER Mathieu VENOT 18 Espace de travail Workspace Variables Charge du robot puissance des moteurs rapport de d multiplication Charge of the robot power motor gear ratio Potential Pour le robot c bles la zone exploitable est arca d finie par la tension et la longueur du fil L valuation de l espace est faite par un 207 m proc d stochastique valu par la m thode des trois forces concurrentes For the wirebot the workable area is defined by the wire tension and length The evaluation of space is made using a stochastic method to determine the reachable space with the method of the three concurrent forces N m SEL a IA Zs ZA LA LEZ AIRS A
16. e the system requires a real time feedback of the construction process in order to control and correct the unexpected behaviour of the aggregates The best way to do this is to use computer vision a branch of weak artificial intelligence that uses sensors to acquire optical data colours shapes depth by means of infrared rays in order to interpret its direct environment The whole system can be divided in four parts Construction Aggregate definition and fabrication Hardware Cable robot operation and instruction sending Software General User Interface and generic shape discretization Artificial intelligence Computer vision and error correction 1 SAUNDERS Peter T DI CRISTINA Giuseppa dir Nonlinearity What It is and why It matters Chichester AD Architecture and Science Wiley Academy 2001 2 DOURSAT R SAYAMA H amp MICHEL O 2013 A review of morphogenetic engineering Frontiers of Natural Computing FNC 2012 Special Issue Lones M Tyrrell A Stepney S amp Caves L eds Natural Computing 12 2 517 535 p 531 3 For an example see Yingchuang New Materials the chinese company that printed 10 houses in a day using large scale concrete printers AABBS Theory Explanation Article Th orique Theorical Paper CONSTRUCTION AGGREGATES Several design and architectural experiments have already dealt with aggregate structures In the past few years serving in fact as the starting
17. eed motor 3 float step s i instruction type integer O 1 2 1 O motors run 1 gripper opening 2 gripper closing 1 motors stop Charles BOUYSSOU Oswald PFEIFFER Mathieu VENOT 26 An alternative to large scale 3D printing AGGREGATED STRUCTURE MODULE CONSTRUCTION Charles BOUYSSOU Oswald PFEIFFER Mathieu VENOT D Aggregated Structure Module Construction Manuel d Utilisateur User Manual Equipement Equipment Programme Software Processing IDE import toxi geom import toxi geom mesh GEOMETRIE import peasy DISPLAY import java awt Frame GUI import java awt BorderLayout GUI import controlP5 GUI import SimpleOpenNI SENSOR import OpenCV VISION MACHINE import processing serial HARDWARE COMMUNICATION Mat riel Hardware Robot ABB 1600 145 pince schunk avec pr hension 2 kinect Xbox ABB Robot 1600 145 ochunk gripper with gripping Kinect 2 Xbox El ments de construction Building elements Module en carton poxy et fibre de verre Lycra Cardboard module Epoxy and fiberglass Lycra R glages Settings Programme Software Les variables sont vitesse de d placement Domaine de vision colorim trique spatiale Domaine de sensibilit a l erreur M thode de choix des modules M thode de choix des cibles a atteindre Variables are Speeding Field of vision color space Sensitiv
18. eory Explanation Article Th orique Theorical Paper so the surrounding targets will be consider as reached already To solve error type 1 the robot needs to detect and grab one particular module for which we use a Blob Detection algorithm allowing the program to identify an isolated object by its outline After checking if this aggregate is reachable if the cables are not going to intersect the structure the robot will be able to grab it and reuse it DEVELOPMENT Of course the AABS detailed here can be seen as a simple proposal to deal with unpredictable material constructions It was never desired that any of the four sub researches that have been presented here be Irreplaceable the aggregates could have different shapes and properties the cable robot could be replaced by drones or by six axis robots the shape discretization could be done with alternative strategies the feedback loop could use other types of sensors etc Diversity Is in fact encouraged as it keeps pushing this approach further The AABS is an interesting proposal because it works in its entirety and therefore proves that it is possible here and now to build complex architectural objects at a large scale based on unpredictable materials and local interactions the user feeds the software some simple rules an overall 3D shape a discretization strategy a variety of aggregates and then the system builds on its own calibrating its actions on a physical fee
19. ity field error Method of selection of modules Method of selection of targets to be achieved Robot Robot Le syst me de positionnement peut tre n importe quel utilitaire de d placement 2 axes ou plus Pour les tests effectu s Robot a cables ABB 1600 145 The positioning system may be any displacement utility with 2 axes or more To perform the tests Cables robot ABB 1600 145 El ments de construction Building elements modules en carton 3 typologies de modules 5 tailles de modules Cardboard module 3 module types 5 module sizes 29 Charles BOUYSSOU Oswald PFEIFFER Mathieu VENOT D Aggregated Structure Module Construction Manuel d Utilisateur User Manual Manuel Manual Robot Robot mise sous tension calibration si n cessaire mise en communication du robot avec l ordinateur Turning trying Calibration if necessary Communication of the robot with the computer Environement Environment Pr paration du rack a modules Modules rack preparation Programme Program Pr paration du model 3D 3D model preparation Construction Construction Mise en route de la communication entre la machine et le programme Initialization of the communication between the machine and the program 31 Adaptative Aggregation Based Building System An alternative to large scale 3D printing AGGREGATED STRUCTURE MODULE EXPLOITATION Plexiglass t
20. point for the broader scope of this project Among others Karola Dierichs and Achim Menges provided a very complete work about the physical behaviour of cohesion based elements at a microscopic scale for their research on aggregate structures conducted at the Institute for Computational Design University of Stuttgart 4 Minimaforms also worked on laser cut and hand projected aggregates for the stage design of a concert during summer 2014 5 Another work that uses clay offers a very similar approach in the machine vision feedback loop and aggregation based construction system Remote material deposition by Gramazio amp Kohler Research 6 As the system exposed in this paper needs to be generic such as most additive manufacturing processes the aggregates shape material and cohesion method can all be variable Just as desktop filament 3D printers were originally developed for basic ABS or PLA but can actually use a large variety of different filaments from tensile plastic to carbon fibre wood or metal based composites Nevertheless for testing purposes and in order to demonstrate the concept of the system the robot needs basic aggregate elements to deal with so a first solution has been developed based on mechanical aggregation The different criteria that have been taken into account so far deal mainly with optimization of the fabrication process producing the highest possible number of modules at the lowest cost and machining assembly time
21. ray Modules Holding skin bedrock Charles BOUYSSOU Oswald PFEIFFER Mathieu VENOT 32 D Aggregated Structure Module Exloitation Manuel d Utilisateur User Manual Equipement Equipment Construction Construction Mat riel Hardware Temps de construction Construction time Empilement de modules mis en forme Epoxy et fibre de verre mise en forme des modules Fondations b ton modules shaping Plateaux Verre PMMA 45 min Module pile shaped Renforcement de la coque Epoxy and fiberglass Shell reinforcement Foundation concrete 2h Boards glass PMMA Ancrage et fabrication de la dalle Slab fabrication and anchoring 2h do
22. tion 8 D finition d une couleur d tecter avec les kinects Definition of a color to track with the kinects 5 Envoi de la simulation Simulation launching 9 D but de la construction Starting the construction 6 Lecture du fichier d analyse Analysis file reading Charles BOUYSSOU Oswald PFEIFFER Mathieu VENOT 12 ONGLET PRINCIPAL ONGLET SIMULATION ONGLET CONSTRUCTION Options de visualisation Options et envoi de la Outils de calibration des d chelle d exports et simulation kinects utilitaire de s lection d acquisition des cibles d une couleur et envoi du code MAIN TAB SIMULATION TAB CONSTRUCTION TAB Visualization scaling export Simulation options and Kinects calibration tool color options and targets acquisition launching tracking utility and code sending Autonomous Aggregation Based System Autonomous Aggregation Based System Autonomous Aggregation Based System ff VISUALIZATIOH f SIMULATIOH ff COHSTRUCTIOH MESH_DISPLAN DISPLAY _BBOH START_STOP_SIM START_STOP_C0HS RETURH_HAIH DISPLAY _EHYIROHMENT ff SETTINGS ff SETTINGS RICQUIRED H RHRLS PARTICULES_OISPLAY CLIPFIHG_FLAHE KIHECT SELECT 90 00 CLIPPIHG_ SIZE 90 00 THRESHOLD_COLOR DISPLAY _HORMALS 15 00 PARTICULES_SIZE COLOR_TRACKED FLIP_HORMALS 200 00 SPEEDSIM SAVE_MATRIH LORD HRTRIH f V XELIZRTIOH 10 00 POIHTS DISTRHCE CALIBRATION TRRHSFORPTS 0 00 TRAHSLATEH_KIMECTI FILL POINTS DISPLAY_FOIHTS ROTATEH KIMEC
Download Pdf Manuals
Related Search