外观
论文arXiv 2610.1797715 天前
ReEnvision：面向长时程操作的未来引导选择性动作修复
原文链接：ReEnvision: Future-Guided Selective Action Repair for Long-Horizon Manipulation（arXiv 2610.17977 · 发布日期：2026-10-13）
ReEnvision: Future-Guided Selective Action Repair for Long-Horizon Manipulation
ReEnvision面向长时程操作偏差，借助未来动作信念选择性修复，在CALVIN、L-CALVIN、RLBench及真机任务验证。
核心要点
- 长时程操作中，局部偏差可能只影响少数动作，整段重规划或逐步纠正各有风险。
- 方法利用执行反馈更新联合未来动作信念，并学习信念修订对未执行动作的影响。
- ReEnvision仅修复缓存动作块中受影响动作，保留有效动作，严重偏差时再完全重规划。
- 在CALVIN、L-CALVIN、RLBench及五个真实操作任务上取得摘要所列成功率提升。
验证范围与局限：验证范围包括CALVIN、L-CALVIN、RLBench和五个真实世界操作任务。
英文摘要 (Abstract)
Long-horizon robotic manipulation requires robust execution over sequences of interdependent actions. Existing methods typically handle execution deviations by replanning the remaining action chunk or correcting actions step by step as new observations arrive. However, a local deviation may affect only a few actions under distribution shift, yet replanning regenerates the entire chunk from an unfamiliar state, potentially introducing errors into otherwise valid actions. Step-wise correction may react too late, as it does not explicitly anticipate how the deviation affects subsequent actions. We propose ReEnvision, a future-guided selective action repair framework that updates a joint future-action belief using execution feedback. It then learns how the resulting belief revision affects each unexecuted action. The resulting belief revision enables selective repair of only the affected actions within a cached action chunk, preserving valid actions and reserving full replanning for severe deviations. On CALVIN, L-CALVIN, and RLBench, ReEnvision achieves 86.6% 5-step success, improves 10-step success from 56% to 63%, and reaches an average success rate of 88.8%, respectively. On five real-world manipulation tasks, it achieves an average success rate of 82.8%. Controlled repair experiments further demonstrate the benefits of localizing affected actions and preserving valid ones during execution.
作者：Xiyin Zeng, Xuewei Huang, Yi Lu, Weihao Yuan, Hao Wang
来源
本页事实对应的公开页面，以原文为准。