Unitree Go2 RL GYM

Forked from Unitree's RL Gym.

Installation

1. Install Isaac Gym

   • Download and install Isaac Gym Preview 4 from https://developer.nvidia.com/isaac-gym
   • cd isaacgym/python && pip install -e .
   • Try running an example cd examples && python 1080_balls_of_solitude.py
   • For extra help:
     • Setting up Isaac Gym
     • Install Isaac Gym

2. Create a new python virtual env with python 3.6, 3.7 or 3.8 (3.8 recommended), you can use the following executable:

   cd isaac gym
   ./create_env_rlgpu.sh
   conda activate rlgpu
   

3. Ensure you have the correct pytorch with cuda for your system. I am using torch==1.13.1+cu117 torchvision==0.14.1+cu117 torchaudio==0.13.1+cu117 with an NVIDIA GeForce RTX 4090 (Notebook):

   pip uninstall torch torchaudio torchvision # if rlgpu's version does not work with your GPU
   pip install torch==1.10.0+cu113 torchvision==0.11.1+cu113 torchaudio==0.10.0+cu113 -f https://download.pytorch.org/whl/cu113/torch_stable.html # recommended 
   

4. Install rsl_rl (PPO implementation)

   • Clone https://github.com/leggedrobotics/rsl_rl
   • cd rsl_rl && git checkout v1.0.2 && pip install -e .

5. Install go2_rl_gym

   • Clone this repository
   • Navigate to the folder go2_rl_gym
   • pip install -e .

Usage (From root directory)

1. Train: python legged_gym/scripts/train.py --task=go2

   • To run on CPU add following arguments: --sim_device=cpu, --rl_device=cpu (sim on CPU and rl on GPU is possible).
   • To run headless (no rendering) add --headless.
   • Important : To improve performance, once the training starts press v to stop the rendering. You can then enable it later to check the progress.
   • The trained policy is saved in logs/<experiment_name>/<date_time>_<run_name>/model_<iteration>.pt. Where <experiment_name> and <run_name> are defined in the train config.
   • The following command line arguments override the values set in the config files:
   • --task TASK: Task name.
   • --resume: Resume training from a checkpoint
   • --experiment_name EXPERIMENT_NAME: Name of the experiment to run or load.
   • --run_name RUN_NAME: Name of the run.
   • --load_run LOAD_RUN: Name of the run to load when resume=True. If -1: will load the last run.
   • --checkpoint CHECKPOINT: Saved model checkpoint number. If -1: will load the last checkpoint.
   • --num_envs NUM_ENVS: Number of environments to create.
   • --seed SEED: Random seed.
   • --max_iterations MAX_ITERATIONS: Maximum number of training iterations.

2. Play:python legged_gym/scripts/play.py --task=go2

   • By default, the loaded policy is the last model of the last run of the experiment folder.
   • Other runs/model iteration can be selected by setting load_run and checkpoint in the train config.