PPO

To run PPO, install the dependencies for each simulator first and run the following command:

isaacgym:

python roboverse_learn/rl/train_rl.py train=CloseBoxPPO sim_name=isaacgym

mujoco:

python roboverse_learn/rl/train_rl.py train=CloseBoxPPO sim_name=mujoco

isaaclab:

python roboverse_learn/rl/train_rl.py train=CloseBoxPPO sim_name=isaaclab

(note: current closebox is set for franka end effector to reach the origin point)

To change PPO configs, check out all files inside roboverse_learn/rl/configs/.