---

Humanoid AMP

Symbolic Links

ln -s ~/IsaacLab/source/isaaclab_tasks/isaaclab_tasks/direct/humanoid_amp ~/IsaacLab/

ln -s ~/IsaacLab/scripts/reinforcement_learning/skrl ~/IsaacLab/

Train

./isaaclab.sh -p ~/IsaacLab/skrl/train.py --task Isaac-G1-AMP-Walk-Direct-v0 --headless

or

./isaaclab.sh -p ~/IsaacLab/skrl/train.py --task Isaac-G1-AMP-Dance-Direct-v0 --headless

Eval

./isaaclab.sh -p ~/IsaacLab/skrl/play.py --task Isaac-G1-AMP-Walk-Direct-v0 --num_envs 32 

TensorBoard

./isaaclab.sh -p -m tensorboard.main --logdir logs/skrl/

Walk training: master branch. Dance training: dance branch.

The usage of some helper script functions is explained at the beginning of the file.

Motions Scripts

• motion_loader.py - Load motion data from npz files and provide sampling functionality
• motion_viewer.py - 3D visualization player for motion data
• data_convert.py - Convert CSV motion data to npz format with interpolation and forward kinematics
• motion_replayer.py - Replay motion data in Isaac Sim with optional recording
• record_data.py - Recording and managing motion data utility classes
• verify_motion.py - Verify and display npz file contents
• visualize_motion.py - Generate interactive HTML charts to visualize motion data
• update_pelvis_data.py - Fix pelvis posture and body center issues in replay motion by replacing pelvis data from source file

Dataset & URDF

Note: The original dataset and URDF files from Unitree Robotics have been removed by the official source.

If you're still looking for the dataset, a third-party mirror is currently available here:
lvhaidong/LAFAN1_Retargeting_Dataset

Or you can also use the data from AMASS

TODO

• ✅ Current: Dancing motion is working
• ✅ Test the test branch thoroughly and merge it into dance as soon as possible
• ✅ Write a more detailed README to cover the new features and usage
• Add clearer comments and explanations in all Python scripts

Resources

Contributions, discussions, and stars are all welcome! ❥(^_-)