论文/arXiv 2610.07525·23 天前

ReDex：通过手指级柔顺交互修复 Sim-to-Real 灵巧操作策略

ReDex通过真人在真机rollout中柔顺纠正单指接触并利用指尖力反馈，适配仿真训练的灵巧策略，并在两项真机任务上验证。

AI 整理

• 硬件零部件
• 本体 SDK 与控制
• 仿真与训练

原文链接：ReDex: Repairing Sim-to-Real Dexterous Policies by Finger-Level Compliant Interaction（arXiv 2610.07525 · 发布日期：2026-10-05）

ReDex: Repairing Sim-to-Real Dexterous Policies by Finger-Level Compliant Interaction

ReDex通过真人在真机rollout中柔顺纠正单指接触并利用指尖力反馈，适配仿真训练的灵巧策略，并在两项真机任务上验证。

核心要点

• 仿真训练的灵巧操作策略常因接触时序和力调节误差导致真实迁移失败。
• ReDex在冻结基策略控制其余手指时，让人类在柔顺控制下纠正选定手指的局部接触失败。
• 方法从rollout重建含力信息的目标，通过行为克隆训练独立的力条件策略。
• 真机实验中，相比基础策略，Object Flipping成功率由14%升至86%，Screwdriver Rotation平均进展由26.0%升至95.3%。

验证范围与局限：验证范围为真实硬件上的两项接触丰富灵巧操作任务，分别覆盖两个和三个物体。

英文摘要 (Abstract)

Dexterous manipulation policies trained in simulation often fail to transfer to the real world because of errors in contact timing and force regulation. Yet these policies can retain useful multi-finger coordination for task progression. We propose ReDex, a framework for adapting a simulation-trained base policy to the real world by correcting local contact failures and incorporating tactile feedback. Starting from a proprioception-only base policy, ReDex allows a human operator to physically correct contact failures at selected fingers under compliant control during real-world rollouts, while the frozen base policy continues to control the remaining fingers. These rollouts combine base policy execution, human-corrected finger motion, and fingertip force observations. We reconstruct force-informed targets from these rollouts to train a standalone force-conditioned policy via behavior cloning. This design reduces human correction effort, enables learning of contact regulation from real-world interaction, and introduces force feedback into a proprioception-only policy without tactile simulation or complex full-hand teleoperation. We evaluate ReDex on two challenging, contact-rich dexterous manipulation tasks on real hardware. Compared with sim-to-real transferred base policies, ReDex increases Object Flipping success rate from 14% to 86% across two objects and average Screwdriver Rotation progress from 26.0% to 95.3% across three objects.

作者：Jinzhou Li, Hadi Tabatabaee, Kelin Yu, Yuyin Sun, Cheng-Hao Kuo, Roberto Martín-Martín, Nima Fazeli, Wu, X. Alice, Xianyi Cheng

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2610.07525：ReDex: Repairing Sim-to-Real Dexterous Policies by Finger-Level Compliant Interaction访问 2026-10-21