HW 1: Out Tues Sep 28, Due Tues Oct 12, 3pm.
Overview
This project has three parts. The first is to explore the difference between reactive agents and model-based agents. The second is to explore the difference between maintaining memory and using cues in the environment. The third is to explore strategic interactions among groups of agents. For each part, you complete skeleton code in java to create a basic agent. Then you evaluate the agent by creating environment specifications that illustrate specific kinds of qualitative behavior. Finally, you have the option to extend the agent in light of your evaluation to give new qualitative behavior, in open-ended ways, for extra credit.
Background to the Project
- HW1 Source An overview of the skeleton code that you will be working with.
- HW1 Specs An overview of the specification language for environment files.
- Advice General advice about coding, testing and debugging.
- Expectations Share your ideas but your work should be your own.
Specific Requirements
- Predators Create reactive and model-based predators that chase other boids.
- Grazers Create reactive and memory-based grazers that move between three different food sources.
- Flockers Create flocks.
Resources
skeleton code files and tarball- example files and tarball of specifications and tarball of sample runs
Updates and changes
FlockingReader.java skeleton code updated 12:25 10/1 to fix window size bug
Simulation.java skeleton code updated 12:25 10/1 to fix window size bug
Flocker.java skeleton code updated 2pm 9/30 to reflect new improved default tuning for weights.
Skeleton.tar updated 12:25 10/1 with current files