具身开发者社区

外观

论文arXiv 2610.2114013 天前

带强化学习先验的用于接触约束感知阻抗控制的 Manifold Generative Annealing ​

提出 MGA 联合无模型与模型生成框架，优化接触任务中的运动与阻抗，并在三类任务上验证。

AI 整理
  • 本体 SDK 与控制
  • 具身模型
  • 评测与安全

原文链接：Manifold Generative Annealing for Contact Constraint-aware Impedance Control with Reinforcement Learning Prior（arXiv 2610.21140 · 发布日期：2026-10-15）

Manifold Generative Annealing for Contact Constraint-aware Impedance Control with Reinforcement Learning Prior

提出 MGA 联合无模型与模型生成框架，优化接触任务中的运动与阻抗，并在三类任务上验证。

核心要点 ​

  • 接触丰富任务需协调运动与阻抗，训练外接触条件可能使学习策略失效。
  • MGA 将预训练强化学习策略组织为结构化序列提案，并保留任务交互策略。
  • 方法通过模型滚动评估提案与探索样本，用于细化或单独选择执行。
  • 实验在标称和感知偏移条件下，提高了阻抗控制的安全成功率。

验证范围与局限：验证范围为表面扫描、插钉和人形推挤任务；摘要未说明仿真或真机。

英文摘要 (Abstract) ​

Contact-rich robotic tasks require coordinated motion and impedance under uncertain interaction dynamics and physical constraints. While recent methods on learned policies provide useful interaction strategies, they may fail when contact conditions differ from those encountered during training. Model-based generative methods show promising receding-horizon trajectories optimization results, but often lack structured behavioral guidance and refinement that jointly accounts for constraint geometry and actual contact responses. To this end, we propose Manifold Generative Annealing (MGA), a joint model-free and model-based generative framework for motion--impedance optimization. MGA first organizes pretrained reinforcement learning policies into structured sequence proposals, retaining task-specific interaction strategies without committing to their execution. It then evaluates these proposals alongside exploratory samples through model-based rollouts for refinement or separate execution selection. Crucially, realization-aware manifold refinement aligns updates from the refinement bank with both task constraints and locally achievable physical responses, allowing learned experience to guide online adaptation under changing contact conditions. Experiments cover surface scanning, peg insertion, and humanoid pushing, with improved safety success rate of impedance control under nominal and sensing-shift conditions.

作者：Zhilin He, 戚国全, Changliu Liu

来源

本页事实对应的公开页面，以原文为准。

  1. arXiv 2610.21140：Manifold Generative Annealing for Contact Constraint-aware Impedance Control with Reinforcement Learning Prior访问 2026-10-21