Abstract

Parkour is a grand challenge for legged locomotion, even for quadruped robots, requiring active perception and various maneuvers to overcome multiple challenging obstacles. Existing methods for humanoid locomotion either optimize a trajectory for a single parkour track or train a reinforcement learning policy only to walk with a significant amount of motion references. In this work, we propose a framework for learning an end-to-end vision-based whole-body-control parkour policy for humanoid robots that overcomes multiple parkour skills without any motion prior. Using the parkour policy, the humanoid robot can jump on a 0.42m platform, leap over hurdles, 0.8m gaps, and much more. It can also run at 1.8m/s in the wild and walk robustly on different terrains. We test our policy in indoor and outdoor environments to demonstrate that it can autonomously select parkour skills while following the rotation command of the joystick. We override the arm actions and show that this framework can easily transfer to humanoid mobile manipulation tasks.

Parkour

Jump Up

Stairs

Tilted Ramp

Jump Down

Leap

Leaping 0.4m and Leaping 0.8m

Slope

Robust Walking

Hovering over the video will play it at 1x speed.

Related Work

This work is based on our Previous Parkour: Robot Parkour Learning


Robot Parkour Learning
Ziwen Zhuang*, Zipeng Fu*, Jianren Wang, Christopher Atkeson, Sören Schwertfeger, Chelsea Finn, Hang Zhao
CoRL 2023


PDF | Video | Project Page

BibTeX

@misc{zhuang2024humanoid,
      title={Humanoid Parkour Learning}, 
      author={Ziwen Zhuang and Shenzhe Yao and Hang Zhao},
      year={2024},
      eprint={2406.10759},
      archivePrefix={arXiv}
}