Introduction to Programming in NetLogo
Contents
1. 4 the Help menu is a great place to start In March 19 2013 ae particular working step by step through the wd chat eee d introduction and first three tutorials see right Setesss sows NetLogo is a programmable modeling environment for simulating p Se System Requirements natural and social phenomena It was authored by Uri Wilensky in will help familiarize you with the programs Contacta Us ro 1999 and has been in continuous development ever since atthe Center for Connected Learning and Computer Based Modeling menus interface commands and possible a NetLogo is particularly well suited for modeling complex systems classroom uses This process will likely take Sample Model Party developing over time Modelers can give instructions to hundreds or thousands of agents all operating independently This makes approximately 3 hours working at a moderate fj teaming Netlog i it possible to explore the connection between the micro level Toona a Conmanda behavior of individuals and the macro level patterns that emerge N etoria 3 Procedures from their interaction pace but is well worth it Some general notes on NetLogo code e The code is organized into blocks that define individual Interface Info Code D routines For example the turtles own block defines key rina i E a properties of each individual agent for historical reasons g rnai eae of nearby turtles NetLogo agents are called turtles Sim2. is a multi agent programmable modeling environment It is used by tens of thousands of students teachers and researchers worldwide It also powers HubNet participatory simulations It is authored by Uri Wilensky and developed at the CCL You can download it free of charge What can you do with NetLogo Read more here Click here to watch videos Join mailing lists here Download NetLogo comes with a large library of sample models Click on some examples below Following installation we recommend that you drag the NetLogo folder from your computer s desktop to your Applications folder Then open NetLogo by double clicking its Desktop icon Open the Models Library from the File menu Choose Flocking from the Biology section and click Open Page 1 of 6 Hot Topics Biological Self Organization Summer 2013 III Interface and Controls In NetLogo the Interface tab lets you interact with and display the model To begin click the setup button and then click go Individual agents should start moving about their simulated world and interacting with each other A few notes e The model will continue to run as long as the go button is selected Click go again to pause or resume the model click setup to restart the model e Use the speed slider to speed up or slow down the model e Use the green slider controls to adjust model parameters Most parameters can be altered even while the model is running however changing
3. the Flocking model for a little while you may want to customize the display add more complex and realistic behaviors or explore other models of self organizing systems NetLogo s built in tutorials user manual and code examples provide many resources to help you do so A few particularly useful features Interface Info Code e You can add plots and controls using the toolbar on the on D Bann becca Interface tab see screenshot at right For example add CELU r A o Switc m4 gt Sticks 179 a switch to turn a particular property on or off or put in a Oi Chaves monitor to output the value of a specific variable within e the model Plots can be customized to display frequency Plot Output histograms or to track one or more numeric variables Fririmumeseparan 7 Note over time using pre built code examples from the k max align turn 5 00 degrees Models Library Histogram Example and Tutorial 3 me respectively n me e The Code Examples section of the Models Library also 300 S contains pre built examples of specific agent behaviors that may be useful when modeling particular biological systems You can copy and paste this code into your own model then modify it as needed For example when an agent in the Flocking model moves past the right hand edge of the simulated world it reappears on the left hand edge 1 e the edges wrap around If you instead want agents in your model
4. Hot Topics Biological Self Organization Summer 2013 Introduction to Programming in NetLogo Anton E Weisstein Truman State University I Background on Agent Based Models An agent based model is a simulation in which individual objects or agents move about and interact with each other and their environment through specified rules The user can fine tune these rules e g by altering agents movement speed or make larger changes e g by creating new kinds of interaction In NetLogo the simulated world is divided into distinct patches Each patch can be given its own appearance properties and rules however the boundaries between patches are usually invisible Similarly a single model may contain multiple kinds or breeds of agent each with its own appearance and behavioral rules In most models agents and patches interact only with other nearby agents and patches Larger scale patterns and behaviors then emerge from these simple localized interactions II Downloading NetLogo and Getting Started NetLogo can be downloaded free of charge from http ccl northwestern edu netlogo The website s Download button will take you to an optional user information form and then to the actual download page The download includes tutorials examples and pre built models as well as the main NetLogo software package Versions of NetLogo are available for Mac OS X Windows and Linux but not currently for phones or tablets NetLogo
5. cs Biological Self Organization Summer 2013 IV The Info Tab In the Display Control see p 2 click the Info tab This will display a text file describing many important aspects of the chosen model including e a general description of the biological system e a summary of the rules governing agents behavior e a description of the model s main controls and parameters e suggested explorations and extensions of the model This information makes the Info tab a logical place to start when introducing a NetLogo model Alternatively you can use the Edit tool see right to remove SS information that you want students to figure out on their own Edited versions of any model should be given a new name using the Save As command from the File menu so that you also retain the original version V Code Tab Programming in NetLogo The Code tab contains the series of commands that determine how the model actually runs This section defines the different kinds of agents and their properties sets up the simulated world in which agents live and lays out agents behavioral rules Understanding and modifying this code allows the user to adapt the model to a much broader range of biological systems However you can also run the standard model and analyze its behavior without ever needing to look at the underlying code If you do want to understand how to code NetLogo in NetLogo the built in User Manual under User Manual version 5 0
6. ilarly the to Marest neighbor 53 closest one of our Flocimates setup block lists the commands to run whenever the user clicks the setup button ae phen crt population set color yellow 2 random 7 set size 1 5 3 easier to see E NetLogo uses color coding to indicate different parts of the setxy random xcor random ycor computer code For example each code block begins _o end with the command to and ends appropriately enough epe with end both of these commands are indicated in green Ae E E ke eed es e en Other commands and mathematical constants are marked animate more smoothly i E repeat 5 ask turtles fd 0 2 display in blue purple or red In addition any text that follows a for greater efficiency at the expense of smooth P E animation substitute the following line instead semicolon is a comment that is not actually part of the ask turtles fd 1 tick code itself but is simply a helpful note for the user niir Comments are automatically colored grey Page 3 of 6 Hot Topics Biological Self Organization Summer 2013 e When you first look at NetLogo code the unfamiliar terminology and formatting may seem daunting However with some practice it becomes fairly easy to deduce the general meaning of many commands As an exercise try to figure out what the following commands from the Flocking module might mean 1 setxy random xcor random y cor Hint This command is part of the to setu
7. ization consider having each pair of students choose one of the above modules read about it through the Info tab and or external resources explore the model s behavior as described in Exploration 5 p 5 and share their findings with the class through a poster session or series of short presentations Page 6 of 6
8. nd merge with different birds emerging as temporary leaders Each bird s position within the flock emerges purely from local interactions between interchangeable agents a defining property of self organizing systems Exploration 5 Working in pairs choose some aspect of the system to study further For example you might explore the effect of a particular variable on the birds behavior or analyze the change in some property over time Write down the question you have chosen to research then take 15 20 minutes to study that question using the model Clearly record your procedures results and inferences Then share your findings with the entire class as a one minute presentation Suggested answer Results will depend on the questions chosen However each group should produce a clearly stated question an appropriate experimental design and a general statement of their results For example students exploring how visual range affects flock size might run the model for values of vision ranging from 0 to 10 patches They might then report that no flocking occurs when vision lt 1 0 patches but that flock size rapidly increases thereafter eventually producing a single large flock when vision approximately 8 patches They might also note that the simulation runs more slowly for larger values of vision because the model must make more calculations as each bird sees more and more of its neighbors VII Extensions Once you have worked with
9. p block that occurs each time the user resets the model 2 repeat 5 ask turtles fd 0 2 display Hint What might fd stand for 3 set flockmates other turtles in radius vision Remember that you can always use NetLogo s built in manual and dictionary if there s something you still don t understand VI Possible Explorations The following set of explorations is intended to help students build from a basic understanding of self organizing systems to an open ended research like project Exploration 1 Run the model for approximately 1000 time steps ticks Describe as clearly as possible how the birds change their flight direction over time What factors directly influence each individual bird s flight direction Suggested answer Each individual bird tends to align its flight direction with those of its neighbors Over time this results in the entire population flying in the same general direction Exploration 2 Reset the model and run it again Do the birds end up flying in the same direction as before Why or why not Suggested answer No Birds initial headings are determined randomly so some headings will be slightly more common than others As each bird aligns with its neighbors those headings will become more and more common eventually becoming the overall direction of the entire population Exploration 3 How many flocks do you see after 1000 time steps Is this number stable or does it change as
10. the model continues to run Note You will have to decide how to determine where one flock ends and another begins make sure that you can explain your reasoning Suggested answer Answers will vary depending on how students interpret the idea of a flock For example some students may decide that a large group of birds represents a single flock whereas other students might say that small gaps within that group justify dividing it into three smaller flocks Some students may focus only on the birds roe while others might also consider their relative headings This can motivate a broader discussion about classification For example how many distinct species do the 15 warblers shown at right represent Screenshot of results from Google Image search for warbler Page 4 of 6 Hot Topics Biological Self Organization Summer 2013 Exploration 4 What special properties if any distinguish flock leaders from other birds in the flock To help answer this question Control click on a single bird at the head of a flock to bring up a menu of options Choose the last option turtle i then select watch turtle i This will create a circle around that bird letting you track its movements more easily To remove the circle Control click on the same bird and choose reset perspective Suggested answer There is no intrinsic difference between flock leaders and other birds As the simulation runs flocks will split a
11. the value of population has no effect until the model is restarted e Many NetLogo models include one or more plots of key variables e g frequency distribution of bird headings average flock size vs time While the original Flocking model lacks any such graphical output NetLogo allows the user to add such features For more details see Section VII Extensions of this guide Display control Controls and settings cae eed between NetLogo s three tabs Allow the user to initialize Speed slider Use the Interface tab to display and control begin and stop the model and Controls how fast the model Info to describe the model and to adjust model parameters the model runs Code to create or edit the model Interface Info Code t F view updates dg h faster on ticks rey eo ee World window 4p ticks 1133 3D population 300 lai Ais a Shows the current state of the simulated world including each agent s location setup go m ae ee vision 3 0 patches E minimum separation 1 00 patches He e max align turn 5 00 degrees T max cohere turn 3 00 degrees z max separate turn 1 50 degrees Distribution of Headings 300 n 2 a Heading Plots and monitors Track and display changes over time not present in original module Command center Allows user to enter text commands directly Command Center observer gt v Page 2 of 6 Hot Topi
12. to bounce like Neighbors Heading Page 5 of 6 Hot Topics Biological Self Organization Summer 2013 billiard balls when they hit the edges use the code from Code Examples gt Bounce Example You can make agents look ahead before moving Look Ahead define different kinds of agents Breed and Shapes and bind together any agents that meet Mobile Aggregation as well as many other behaviors In general if there s a particular behavior you want in your model there s a good chance that one of the pre built models already has the code you need Because bird flocking is so familiar it serves as a good starting point from which to introduce the concept of self organization However self organizing systems are not restricted to the organismal scale or even to biology The following NetLogo models demonstrate self organization in a wide variety of settings Biology gt Ants Fireflies Fur Heatbugs Membrane Formation Moths Slime Termites Chemistry amp Physics gt Crystallization gt Crystallization Basic Chemistry amp Physics gt Diffusion Limited Aggregation gt DLA Simple Chemistry amp Physics gt Polymer Dynamics Computer Science gt Cellular Automata gt Life Earth Science gt Erosion Fire Mathematics gt Voronoi Emergent Networks gt Giant Component Team Assembly Social Science gt Segregation Traffic Basic To help communicate the shared features of self organ
Download Pdf Manuals
Related Search