Reproduce “Body Schema Policy: Body as the Vision-Action Interface for Efficient Robot Loco-Manipulation” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Lechen Zhang, Chenyuan Chen, Muxin Zhou, Zewen He, Yoshihiko Nakamura

The Body Schema Policy (BSP) organizes imitation-learning features around robot joints and links, allowing body parts to query image patches while a policy refines action chunks. On nine Unitree G1 SIMPLE tasks, BSP achieved 221 successes out of 270, versus 220 for Ψ₀, with 35.0M parameters and 13.9-millisecond latency. These results are limited to benchmark episodes.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.