面向冲击感知灵巧接物的结果敏感运动搜索

中文摘要 AI 辅助

熟练的人类能够通过协调拦截、速度匹配和随球动作来柔和地接住快速移动的物体，从而减轻冲击。然而，使用强化学习（RL）学习这种冲击感知的接物行为具有挑战性，因为策略必须在调节敏感的接触过渡的同时实现可靠的拦截和抓取。此外，即使是一个能力强大的特权状态RL教师也可能无法为可部署的模仿学习（IL）学生提供理想的示范：教师的失败限制了任务覆盖范围，而接触前运动的微小变化可能产生显著不同的冲击和抓取结果。我们通过干预结果敏感性来表征这一现象，并引入结果敏感窗口（OSW）来指导有针对性的示范构建。基于这一表述，我们提出了结果敏感运动搜索，它学习一个成功的OSW运动的任务条件流形，并执行局部测地搜索来改进成功的教师轨迹，并修复教师失败的任务条件。然后，我们在校准的IL学生动作误差模型下通过完整轨迹验证候选运动，并仅保留成功的执行作为示范。广泛的模拟实验表明，我们的方法有效地修复了教师失败的任务条件，并使由此产生的IL策略在接物成功率和冲击减轻方面均优于特权RL教师。

英文摘要

Skilled humans can catch fast-moving objects softly by coordinating interception, velocity matching, and follow-through to mitigate impact. Learning such impact-aware catching with reinforcement learning (RL), however, is challenging, as the policy must achieve reliable interception and grasping while regulating the sensitive transition into contact. Moreover, even a capable privileged-state RL teacher may not provide ideal demonstrations for a deployable imitation-learning (IL) student: teacher failures limit task coverage, while small variations in pre-contact motion can produce substantially different impact and grasping outcomes. We characterize this phenomenon through interventional outcome sensitivity and introduce the outcome-sensitive window (OSW) to guide targeted demonstration construction. Building on this formulation, we propose Outcome-Sensitive Motion Search, which learns a task-conditioned manifold of successful OSW motions and performs local geodesic search to refine successful teacher rollouts and repair task conditions where the teacher fails. We then validate candidate motions through complete rollouts under a calibrated IL-student action-error model and retain only successful executions as demonstrations. Extensive simulation experiments demonstrate that our method effectively repairs task conditions where the teacher fails and enables the resulting IL policy to outperform the privileged RL teacher in both catching success and impact mitigation.

发表机构

• College of Robotics Science and Engineering, Taiyuan University of Technology（太原理工大学机器人科学与工程学院）
• Department of Mechanical Engineering, The Hong Kong Polytechnic University（香港理工大学机械工程学系）
• Southern University of Science and Technology（南方科技大学）
• School of Advanced Engineering, Great Bay University（大湾区大学先进工程学院）
• College of Mechanical and Electrical Engineering, Northeast Forestry University（东北林业大学机电工程学院）
• Department of Production Engineering, KTH Royal Institute of Technology（瑞典皇家理工学院生产工程系）
• Kaiyang Laboratory, Chery Automobile Co., Ltd.（奇瑞汽车股份有限公司开阳实验室）

机构由 AI 辅助整理，请以论文原文为准。