[论文] FACTR 2: Learning External Force Sensing for Commodity Robot Arms...

论文概要

研究领域: ML 作者: Steven Oh, Jason Jingzhou Liu, Tony Tao, Philip Han, Kenneth Shaw, Satoshi Funabashi, Ruslan Salakhutdinov, Deepak Pathak 发布时间: 2026-06-10 arXiv: 2606.12406

中文摘要

接触丰富的操作需要力敏感性，但许多机械臂因高成本缺乏专用力传感器。我们提出神经外部扭矩估计（NEXT），一种数据驱动方法，无需任何专用力传感器即可估计外部关节扭矩。NEXT仅使用10分钟自由运动数据，1分钟即可完成训练，却能达到与专用关节扭矩传感器相当的估计精度。NEXT使低成本机械臂的力反馈遥操作成为可能，并通过力知情重采样训练（FIRST）改善策略学习，FIRST在行为克隆中对接触前和接触段进行上采样。在五个长期任务中，FIRST比之前力感知策略的任务进度提升超过17%。NEXT和FIRST共同为现成机器人带来力感知遥操作和策略学习，无需额外传感硬件。

原文摘要

Contact-rich manipulation requires force sensitivity, but many robot arms lack dedicated force sensors due to their high cost. We present Neural External Torque Estimation (NEXT), a data-driven method that estimates external joint torques without needing any dedicated force sensors. NEXT trains in 1 minute from only 10 minutes of free-motion data, yet achieves estimates comparable to dedicated joint-torque sensors. NEXT enables force-feedback teleoperation on low-cost arms and improves policy learning through Force-Informed Re-Sampling Training (FIRST), which up-samples pre-contact and contact segments during behavior cloning. Across five long-horizon tasks, FIRST outperforms prior force-aware policies by over 17% in task progress. Together, NEXT and FIRST bring force-aware teleoperation a...

--- *自动采集于 2026-06-12*

#论文 #arXiv #ML #小凯