通过接触信赖域实现灵巧的富接触操作
Dexterous Contact-Rich Manipulation via the Contact Trust Region
AI总结：
针对富接触操作中接触动力学局部描述与单边性质不一致的问题，提出 Contact Trust Region (CTR)，据此开发局部 MPC 算法并扩展至全局规划，在仿真与硬件实验中以更低计算成本实现了高效的灵巧操作。
摘要：
什么是富接触操作中接触动力学的良好局部描述，我们在哪里可以信任这种局部描述？虽然许多方法通常依赖动力学的 Taylor 逼近及其椭球信赖域，但我们认为此类方法与接触的单边性质存在根本上的不一致。作为补救，我们提出了 Contact Trust Region (CTR)，它既能捕捉接触的单边性质，又能保持计算的高效性。借助 CTR，我们首先开发了一种能够合成局部富接触规划的 Model-Predictive Control (MPC) 算法。随后，我们通过拼接局部 MPC 规划将此能力扩展至全局规划，从而实现高效且灵巧的富接触操作。为验证本方法的性能，我们在两个富接触系统上进行了全面评估：包括高保真仿真和硬件测试，这两个系统分别为平面 IiwaBimanual 系统和 3D AllegroHand 系统。在这两个系统上，与现有的基于 RL 的富接触操作方法相比，我们的方法提供了一种计算量显著更低的替代方案。具体而言，我们以 roadmap 形式呈现的 Allegro 灵巧手内操作策略，在一台标准笔记本电脑上仅使用 CPU 即可在不到 10 分钟内完成离线构建，而在线推理仅需几秒钟。实验数据、视频和代码可在 ctr.theaiinstitute.com 获取。
英文摘要：
What is a good local description of contact dynamics for contact-rich manipulation, and where can we trust this local description? While many approaches often rely on the Taylor approximation of dynamics with an ellipsoidal trust region, we argue that such approaches are fundamentally inconsistent with the unilateral nature of contact. As a remedy, we present the Contact Trust Region (CTR), which captures the unilateral nature of contact while remaining efficient for computation. With CTR, we first develop a Model-Predictive Control (MPC) algorithm capable of synthesizing local contact-rich plans. Then, we extend this capability to plan globally by stitching together local MPC plans, enabling efficient and dexterous contact-rich manipulation. To verify the performance of our method, we perform comprehensive evaluations, both in high-fidelity simulation and on hardware, on two contact-rich systems: a planar IiwaBimanual system and a 3D AllegroHand system. On both systems, our method offers a significantly lower-compute alternative to existing RL-based approaches to contact-rich manipulation. In particular, our Allegro in-hand manipulation policy, in the form of a roadmap, takes fewer than 10 minutes to build offline on a standard laptop using just its CPU, with online inference taking just a few seconds. Experiment data, video and code are available at ctr.theaiinstitute.com.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
