Search papers, labs, and topics across Lattice.
This paper introduces Pose Flow Matching for Humanoid Robots (PFM-HR), a novel approach that leverages a reusable flow matching prior trained on large-scale unordered pose data to enhance reinforcement learning in humanoid tracking. By employing the Pose Geometry Score (PGS), which assesses the alignment of joint coordinate changes with local pose geometry, the method effectively modulates tracking rewards to facilitate structured pose transitions during policy exploration. Experimental results indicate that PFM-HR significantly improves tracking performance for both single and general motion tasks, particularly in dynamic scenarios.
A new flow matching prior can drastically enhance humanoid motion tracking by guiding policy exploration with geometric insights from unordered pose data.
Motion priors improve reinforcement learning for physics-based humanoid tracking, but temporal priors require ordered motion clips, while pose priors provide limited guidance for policy-induced pose transitions. We present Pose Flow Matching for Humanoid Robots (PFM-HR), a reusable flow matching prior trained directly on large scale unordered pose data. PFM-HR introduces the Pose Geometry Score (PGS), which quantifies how joint coordinate changes during rollouts align with the local geometry of pose variation captured by the prior. Using PGS to modulate the tracking reward guides policy exploration toward structured pose changes while keeping the prior frozen across tracking tasks. Experiments demonstrate that PFM-HR improves both single motion and general motion tracking, especially for highly dynamic motions.
