OmniTrack: General Motion Tracking via Physics-Consistent Reference

TL;DR

OmniTrack is a two-stage framework that improves humanoid robot motion tracking by generating physically feasible reference motions and training control policies to follow them, enabling stable and generalizable behaviors.

Contribution

The paper introduces OmniTrack, a novel approach that decouples physical feasibility from motion tracking, enhancing stability and generalization in humanoid robot control.

Findings

Improved tracking accuracy over baseline methods.

Successful real-world execution of complex motions like flips.

Supports human-style teleoperation with robustness.

Abstract

Learning motion tracking from rich human motion data is a foundational task for achieving general control in humanoid robots, enabling them to perform diverse behaviors. However, discrepancies in morphology and dynamics between humans and robots, combined with data noise, introduce physically infeasible artifacts in reference motions, such as floating and penetration. During both training and execution, these artifacts create a conflict between following inaccurate reference motions and maintaining the robot's stability, hindering the development of a generalizable motion tracking policy. To address these challenges, we introduce OmniTrack, a general tracking framework that explicitly decouples physical feasibility from general motion tracking. In the first stage, a privileged generalist policy generates physically plausible motions that strictly adhere to the robot's dynamics via trajectory rollout in simulation. In the second stage, the general control policy is trained to track these physically feasible motions, ensuring stable and coherent control transfer to the real robot. Experiments show that OmniTrack improves tracking accuracy and demonstrates strong generalization to unseen motions. In real-world tests, OmniTrack achieves hour-long, consistent, and stable tracking, including complex acrobatic motions such as flips and cartwheels. Additionally, we show that OmniTrack supports human-style stable and dynamic online teleoperation, highlighting its robustness and adaptability to varying user inputs.