A gym reinforcement learning environment for OpenCat robots based on Stable-Baselines3 and PyBullet.
Start training with train_nybble.py or train_bittle.py To take a look at the pre-trained example, execute enjoy_nybble.py or enjoy_bittle.py.
For more information on the reinforcement training implementation: https://stable-baselines3.readthedocs.io/en/master/index.html And for the simulation environment please refer to: https://pybullet.org/wordpress/
# M1Mac に Miniforge をインストール (Ventura 13.2.1を使用)
brew install miniforge
# 仮想環境の作成
conda env create --name opencat --file requirements.yaml
# 仮想環境のアクティベート
conda activate opencat
# サンプルコード実行
python train_bittle.py
# 仮想環境の終了
conda deactivate