Spatial-Temporal Motion Retargeting (Control policy Learning)

1. Install pytorch 1.13 with cuda-11.6: pip install torch==1.13.1+cu116 torchvision==0.14.1+cu116 torchaudio==0.13.1 --extra-index-url https://download.pytorch.org/whl/cu116
2. Install Isaac Gym
   • Download and install Isaac Gym Preview 3 (Preview 2 will not work!) 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 troubleshooting check docs isaacgym/docs/index.html)
3. Install rsl_rl (PPO implementation)
   • Clone this repository
   • cd AMP_for_hardware/rsl_rl && pip install -e .
4. Install legged_gym
   • cd ../ && pip install -e .

Generating referece data

git clone both and this repositories in the same directory

1. Spatial Motion Retargeting (SMR): https://github.com/terry97-guel/Quadruped_Retargeting
2. Temporal Motion Retargeting (TMR): https://github.com/terry97-guel/Quadruped-Motion-Timing

Usage in simulation

1. Train: python legged_gym/scripts/train.py --task={ROBOT}_{MR}_{MOTION}
   • ROBOT: go1, go1base, a1, al (i.e. aliengo)
   • MR: NMR, SMR, TMR, STMR, TO
   • MOTION: trot0, trot1, go1trot, pace0, pace1, hopturn, sidesteps, videowalk0, videowalk1
2. Play target motion: python legged_gym/scripts/play_target_motion.py --task={ROBOT}_{MR}_{MOTION}
3. Play policy python legged_gym/scripts/play.py --task={ROBOT}_{MR}_{MOTION}
   • This also exports policy to the folder named "export".

Real-world deployment

1. connect to go1 sh {REPO_PATH}/go1_gym_deploy/scripts/go1_connect.sh
2. ssh to go1 system ssh [email protected]
3. run position controller node sudo ~/go1_gym/go1_gym_deploy/unitree_legged_sdk_bin/lcm_position
4. start docker node cd ~/go1_gym/go1_gym_deploy/docker && sudo make autostart
5. run deploy script python go1_gym_deploy/scripts/deploy_policy.py

License

This repository and its code are referred from

1. https://github.com/Alescontrela/AMP_for_hardware
2. https://github.com/Improbable-AI/walk-these-ways
3. https://github.com/Denys88/rl_games