Ant_racer,

a virtual multi-agent pursuit-evasion platform based on OpenAI/Gym and Mujoco

This project is a variant of multi-agent game platform TripleSumo (Publication, Repository). A live demo of this game can be found in This Video. You're welcome to visit the author's Youtube page to find more about her work. Contact her at [email protected] if you have inquiry.

Steps of installing Ant_racer:

1. Download Mujoco200, rename the package into mujoco200, then extract it in /home/your_username/.mujoco/ , then download the license into the same directory
2. Add export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/home/your_username/.mujoco/mujoco200/bin to your ~/.bashrc, and then source ~/.bashrc
3. Use Anaconda to create a virtual environment 'ant_racer' with conda env create -f ant_racer_env.yml; Then conda activate ant_racer.
4. git clone https://github.com/niart/Ant_racer.git and cd Ant_racer
5. Use the gym foler of this repository to replace the gym installed in your conda environment ant_racer.
6. To test the demo, run python chase_demo.py. If you meet error Creating window glfw ... ERROR: GLEW initalization error: Missing GL version, you may add export LD_PRELOAD=/usr/lib/x86_64-linux-gnu/libGLEW.so to ~/.bashrc, then source ~/.bashrc.

A simple RL algorithm interface has been written in chase_runmanin.py which implement DDPG. Important training steps are in /gym/envs/mujoco/chase.py

To cite this platform:

@misc{Ant_racer,
  howpublished = {\href{https://github.com/niart/Ant_racer}{N. Wang, Ant_racer: a multi-agent pursuit-evasion platform. Github Repository, 2021, https://github.com/niart/Ant_racer}},} 

An overview of Ant_racer game: