This project focuses on training the Ant-v4 quadruped environment in MuJoCo using several reinforcement learning algorithms provided by Stable-Baselines3. The algorithms used in this repository are:
- Deep Deterministic Policy Gradient (DDPG)
- Soft Actor-Critic (SAC)
- Twin Delayed Deep Deterministic Policy Gradient (TD3)
- Proximal Policy Optimization (PPO)
- Advantage Actor-Critic (A2C)
- Python Version: 3.7.10
- Stable-Baselines3 Version: 1.5.1a8
- PyTorch Version: 1.11.0
- GPU Enabled: True
- Numpy Version: 1.21.2
- Gym Version: 0.21.0
The repository includes the training scripts, model definitions, and environment configurations required to implement the algorithms listed above. The Ant-v4 environment is used to simulate the movement and control of a quadruped robot in a MuJoCo simulation environment.
