Multi-Agent Interface

Multi-Agent Interface#


PettingZoo ( is a collection of multi-agent environments for reinforcement learning. We build a pettingzoo interface for flatland.


PettingZoo is a popular multi-agent environment library ( that aims to be the gym standard for Multi-Agent Reinforcement Learning. We list the below advantages that make it suitable for use with flatland

act = model.predict(obs, deterministic=True)[0] 
  • Parallel learning using literally 2 lines of code to use with stable baselines 3

env = ss.pettingzoo_env_to_vec_env_v0(env)
env = ss.concat_vec_envs_v0(env, 8, num_cpus=4, base_class=stable_baselines3)
  • Tested and supports various multi-agent environments with many agents comparable to flatland. e.g.

  • Clean interface means we can custom add an experimenting tool like wandb and have full flexibility to save information we want