Running head: Boids Simulation

Boids Simulation
Kelon Grandberry
BS at University of Tennessee at Martin

Assignment #1 Submitted in Partial Fulfillment of the Requirements for the Course
CIS 530: Simulation & Modeling I

Dr. Glenn Hines
Strayer University
Winter 2012

Contents Kelon Grandberry 3 Date: January 28, 2011 3 Abstract 4 Two examples of a boid simulation. For each one, find a description of the rules that each boid follows. Compare and contrast the two sets of rules 5 Are the boids flying in a 2-D or 3-D space? Are they flying in a world with boundary or without boundary? Describe the world or universe in which the boids are flying ………….….6 Does the model have user-adjustable parameter? Such parameters include the number of boids, the closeness factor (how close one boid must one boid must be to the other boids before it’s affected by the other’s behavior), speed and so forth: Make a list of such parameters………………………………………………………………………………………………………………………………..6 How long does it take for flocking behavior to emerge………………………………………………….…………7 References 9

Certificate of Authorship

I Kelon Grandberry certify that I am the author of this document and any assistance I received in preparing this report fully acknowledged. I have also cited in APA format for all sources that I obtained ideas, data, and words. Sources are properly credited according to the APA guidelines.
Kelon Grandberry
Date: January 28, 2011

Abstract
Assignment 1 will be an overview of what was discussed in chapter 4; the title of this paper is called the “Boids Simulation”. This paper will address two examples of boid simulation also discuss their particular rules and compare and contrast the two simulations. This paper explores boids in a 2d or 3d space depending on the experiment. Also, are the boids flying in a world with boundary or without boundary? This paper also describes the world or universe in which the boids are flying or living in. motion of a flock of birds, and a school of fish which is a familiar part of the world we live in. This paper addresses user-adjustable parameters we have to figure out the number of boids, the closeness factor, speed and so forth. Finally, this paper addresses the length for flocking behavior to emerge.

1) Two examples of a boids simulation. For each one, find a description of the rules that each boid follows. Compare and contrast the two sets of rules.
After searching through various simulations online websites I’ve found two examples that were very interesting in my opinion. My first example of a boid simulation is from the Tim Burton film Batman returns. This film was the first to contain computer generated flocks and swarms. It contained computer simulated bat swarms and penguin flocks which were created with the original boid software developed. (Reynolds, 1986) My next example of a boid simulation is more of a realistic type of example and that is working with a flock of ants. For the most part each of the boids go through the same procedure as any other boid simulation. The boid flocks on the basis of three simple rules of interaction: collision avoidance, velocity matching or alignment, and flock centering. (Reynolds, 1986)
Watching the batman scene with the penguins and also with the bats all together made me believe that the computer animator followed the separation definition because you always saw the animal’s kind of scatter away only because they didn’t want to draw contact with each other or any static objects. As a kid, I can remember putting candy on the ground and watching the ants all pile together. Now this experiment, is similar to the batman experiment in many ways, but I think the difference between the batman example and the ant example is the flock of ants always seemed to keep a dynamic equilibrium, they never separated from one another and then came back were as the batman example I saw the bat’s leave away from the flock of other bats only to eventually come back.

2) Are the boids flying in a 2-D or 3-D space? Are they flying in a world with boundary or without boundary? Describe the world or universe in which the boids are flying.
The batman returns example is flying in a 3d space with without any boundaries at all. The bats and also the penguins in that movie did not have a designated leader, and none of them really know who is going the right direction. Most of the bats are just massed in a boids-like swarm, following rules to fly with their neighbors. Since the bats are going slightly faster, they eventually reach the front and have to change direction back within the swarm for a moment. But with most of the bats just trying to follow basic rules and stay in the swarm, the few who know where they are going are good enough: with just a small number who are flying quickly in a fixed direction, the entire group will end up in the right place. I think that the bats and penguins all practiced the collision avoidance rule. Collision avoidance means that the boids should not fly into other boids, so there is a kind of repulsion from other objects. Velocity matching means that the boids adjust their velocity to match the average velocity their neighbors: it will speed up or slow down depending on what the others are doing. (Pfier, Bongard, Berry; 2011, p.58)
The ant example is more ground oriented instead of flying in the air like birds or bats. This particular boid is in a 2D space and is in a world with boundary. Unlike the previous example the ant has a goal to accomplish. Although they still follow the collision avoidance rule they still are trying to reach the food that was dropped on the ground which is a common goal for all ants that are in the model. They seem to use the swarm intelligence application which solves combinatorial optimization problems by intimidating the mechanisms used by ants for path-optimization. Their world seems to go a lot slower than it does with bat, birds, or anything that’s flying in the air.
3) Does the model have user-adjustable parameter? Such parameters include the number of boids, the closeness factor (how close one boid must one boid must be to the other boids before it’s affected by the other’s behavior), speed and so forth: Make a list of such parameters. According to Sichman & Antunes (2006): Depending on the motivation of the modeling work, the criteria used to explore the parameter space will also be different. This motivation may be to model and simulate a real system, but it can also be to study the discrete model that may produce a given emergent phenomenon. Finally, the motivation may be to propose models that perform best in the realization of a specific function. (p.43)
We will present the parameter setting of an agent-based model with the example of ant foraging. The principle is that each ant searches for food and brings it to the nest secreting a chemical on the way up to the food source, which reinforces the presence of the chemical and finally produces trails between the nest and the food sources. We can see ant lines emerging, which are similar to the ones that we can observe in natural conditions. In the model that we used, there is a nest in the center of the area of the simulation, and three food sources around the nest. (p.43)
In this model, two parameters condition the formation of chemical trails. The first one is the diffusion rate of the chemical, which correspond to the fact that a given proportion of the chemical will be diffused to the neighboring patches (regions of the environment) at the next time step. This is used to simulate the diffusion of the chemical in the atmosphere. The second parameter is given the evaporation rate of the chemical, which corresponds to the fact that a given proportion of the chemical will disappear from the patch at the next step. This is used to simulate the evaporation of the chemical in the atmosphere. (p.44)

4) How long does it take for flocking behavior to emerge?
According to Grimm & Railsback (2005): Boids has become a classic illustration of emergence. The flocking behavior of a system meets the criteria for emergence stated in section 5.2: flocking behavior is a system-level behavior, not just the sum of individual properties, and flocking behavior is of a different type (spatial pattern) than individual properties (individuals cannot have spatial patterns, only locations). Most important, although we might anticipate some kind of flocking behavior from the three rules individual boids use to navigate, many characteristics of the flock’s behavior cannot be predicted just by looking at the individual-level rules. In fact, the flock emergent behavior depends on the number of boids and the characteristics of the space as well as the trait of the boids.

References
Grim, V., Railsback, S.(2005). Individual –based Modeling and Ecology.Princeton,NJ:Princeton University Press.
Pfier, R., Bongard,J.,Berry,D.(2011). Designing Intelligence Why Brains Aren’t Enough. Norderstedt Germany: Grin Verlag
Reynolds, C. (1987). Boids Background and Update.red3d.com Retrieved February 2, 2012 from http://www.red3d.com/cwr/boids/
Sichman, L., Antunes, L. (2006). Multi-Agent-Based Simulation IV. Germany: Springer Science Business Media.