Abstract:Flow-matching robot policies commonly use action-chunking inference for efficient closed-loop control, but chunk boundaries can introduce discontinuous action transitions. Existing RTC guidance improves continuity by injecting correction signals during denoising, yet its weight schedule is weak at intermediate timesteps and its unconstrained correction direction may introduce transverse perturbations. We propose POTR, a **p**rior-corrected **o**rthogonal **t**rust-**r**egion guidance method. First, we incorporate a data-prior scale $\sigma_d$ into the RTC guidance weight, yielding stronger intermediate-time correction. Second, we decompose the guidance vector into components parallel and perpendicular to the denoising velocity, and constrain the perpendicular component within a trust region. On LIBERO with $\pi_{0.5}$, POTR improves success rate and consistently reduces chunk-boundary discontinuity, acceleration, and jerk compared with RTC. Ablations show that the prior-corrected weight provides the main correction gain, while the orthogonal trust region further improves stability.
| Subjects: | Robotics (cs.RO); Machine Learning (cs.LG) |
| Cite as: | arXiv:2605.24433 [cs.RO] |
| (or arXiv:2605.24433v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2605.24433 arXiv-issued DOI via DataCite (pending registration) |
From: Kai Fang [view email]
[v1]
Sat, 23 May 2026 06:59:13 UTC (3,724 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。