发表机构
Mitsubishi Electric Research Labs（三菱电机研究实验室）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
ContactDP提出一种多模态扩散策略框架，融合视觉、触觉与力觉信息，结合混合位置-力控制，在紧密插入任务中显著提升性能、可靠性与泛化能力。
AI 中文摘要
高精度连接器插入对机器人系统而言仍具挑战性，原因在于严苛的机械公差、接触过程中的部分可观测性，以及由遮挡和接触模糊性引起的多模态不确定性。成功的插入需要闭环接触引导，该引导需持续整合全局对齐线索与局部接触反馈，以在交互作用下产生稳定的纠正动作。在本工作中，我们提出了ContactDP（面向紧密插入任务的接触引导扩散策略），一种用于接触丰富插入任务的多模态扩散策略框架。ContactDP联合整合了腕部RGB观测、指尖触觉感知和腕部安装的力-力矩测量，以推断接触状态并在插入过程中生成时间上一致的纠正运动。为确保接触下的稳定执行，学习到的策略与一个混合位置-力控制器协同工作，该控制器提供柔顺的低层交互。我们在多种工业级连接器插入任务上评估了我们的方法，这些任务具有不同的连接器几何形状、抓取条件和初始错位。在所有任务中，ContactDP在性能、可靠性和泛化性方面显著优于仅视觉的扩散策略。
英文摘要
High-precision connector insertion remains challenging for robotic systems due to tight mechanical tolerances, partial observability during contact, and multimodal uncertainty arising from occlusion and contact ambiguity. Successful insertion requires closed-loop contact guidance that continuously integrates global alignment cues with local contact feedback to produce stable corrective actions under interaction. In this work, we present ContactDP (Contact-Guided Diffusion Policy for Tight Insertion Tasks), a multimodal diffusion-policy framework for contact-rich insertion. ContactDP jointly integrates wrist RGB observations, fingertip tactile sensing, and wrist-mounted force-torque measurements to infer contact state and generate temporally consistent corrective motions during insertion. To ensure stable execution under contact, the learned policy operates together with a hybrid position-force controller that provides compliant low-level interaction. We evaluate our approach on a suite of industrial-grade connector insertion tasks with varying connector geometries, grasp conditions, and initial misalignment. Across all tasks, ContactDP significantly outperforms vision-only diffusion policies for performance, reliability and generalization.
Comments8 pages, 5 figures
