Robot Parkour Learning

Robot Parkour Learning

Project website: robot-parkour.github.iohttps://robot-parkour.github.io/ (opens in a new tab)

Authors: Ziwen Zhuang* (opens in a new tab), Zipeng Fu* (opens in a new tab), Jianren Wang (opens in a new tab), Christopher Atkeson (opens in a new tab), Sören Schwertfeger (opens in a new tab), Chelsea Finn (opens in a new tab), Hang Zhao (opens in a new tab)

Conference on Robot Learning (CoRL) 2023, Oral, Best Systems Paper Award Finalist (top 3)

Repository Structure

• legged_gym: contains the isaacgym environment and config files.
  • legged_gym/legged_gym/envs/{robot}/: contains all the training config files for a specific robot
  • legged_gym/legged_gym/envs/base/: contains all the environment implementation.
  • legged_gym/legged_gym/utils/terrain/: contains the terrain generation code.
• rsl_rl: contains the network module and algorithm implementation. You can copy this folder directly to your robot.
  • rsl_rl/rsl_rl/algorithms/: contains the algorithm implementation.
  • rsl_rl/rsl_rl/modules/: contains the network module implementation.

Training in Simulation

To install and run the code for training A1/Go2 in simulation, please clone this repository and follow the instructions in legged_gym/README.md (opens in a new tab).

Hardware Deployment

To deploy the trained model on your unitree Go1 robot, please follow the instructions in Deploy-Go1.md (opens in a new tab) for deploying on the Unittree Go1 robot.

To deploy the trained model on your unitree Go2 robot, please follow the instructions in Deploy-Go2.md (opens in a new tab) for deploying on the Unittree Go2 robot.

Trouble Shooting

If you cannot run the distillation part or all graphics computing goes to GPU 0 dispite you have multiple GPUs and have set the CUDA_VISIBLE_DEVICES, please use docker to isolate each GPU.

NOTE: When playing the distilled policy in simulation, DO remember to set ckpt_manipulator = None when configuring the runner. Please refer to #82 (opens in a new tab) for more detail.

To Do

• Go1 training configuration (does not guarantee the same performance as the paper)
• A1 deployment code
• Go1 deployment code
• Go2 training configuration example (does not guarantee the same performance as the paper)
• Go2 deployment code example

Citation

If you find this project helpful to your research, please consider cite us! This is really important to us.

@inproceedings{
    zhuang2023robot,
    title={Robot Parkour Learning},
    author={Ziwen Zhuang and Zipeng Fu and Jianren Wang and Christopher G Atkeson and S{\"o}ren Schwertfeger and Chelsea Finn and Hang Zhao},
    booktitle={Conference on Robot Learning {CoRL}},
    year={2023}
}