ROLGJun 15, 2025

RL from Physical Feedback: Aligning Large Motion Models with Humanoid Control

arXiv:2506.12769v113 citationsh-index: 11
Originality Incremental advance
AI Analysis

It addresses a critical sim-to-real gap in robotics for efficient learning of new behaviors, though it is incremental as it builds on existing text-to-motion generation methods.

This paper tackles the problem of generating physically feasible motions for humanoid robots from text instructions, which existing methods often fail to do, and proposes RLPF, a framework that integrates physics-aware evaluation and text-conditioned generation to achieve both physical plausibility and semantic alignment, enabling successful real-world deployment.

This paper focuses on a critical challenge in robotics: translating text-driven human motions into executable actions for humanoid robots, enabling efficient and cost-effective learning of new behaviors. While existing text-to-motion generation methods achieve semantic alignment between language and motion, they often produce kinematically or physically infeasible motions unsuitable for real-world deployment. To bridge this sim-to-real gap, we propose Reinforcement Learning from Physical Feedback (RLPF), a novel framework that integrates physics-aware motion evaluation with text-conditioned motion generation. RLPF employs a motion tracking policy to assess feasibility in a physics simulator, generating rewards for fine-tuning the motion generator. Furthermore, RLPF introduces an alignment verification module to preserve semantic fidelity to text instructions. This joint optimization ensures both physical plausibility and instruction alignment. Extensive experiments show that RLPF greatly outperforms baseline methods in generating physically feasible motions while maintaining semantic correspondence with text instruction, enabling successful deployment on real humanoid robots.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes