用力觉反馈提升机器人抓取泛化能力，训练时逐步减少视觉干扰。
FACTR: Force-Attending Curriculum Training for Contact-Rich Policy Learning
- 通过渐进式视觉干扰课程，引导模型关注力觉信息。
- 在未见物体上泛化性能比基线提升43%。
- 适合需要精细力控的复杂操作任务研究者。
人类执行许多接触丰富的任务（如搬箱子或擀面团）依赖力觉反馈以确保可靠执行。然而，尽管大多数机械臂都具备力觉信息，该信息在遥操作和策略学习中却很少被利用。因此，机器人的行为通常局限于无需复杂力反馈的准静态运动任务。本文首先提出一种低成本、直观的双向遥操作装置，将从动臂的外部受力反馈至主臂，从而支持复杂接触任务的数据采集。随后引入FACTR方法，采用一种课程学习策略，逐步降低训练过程中对视觉输入的干扰强度。该策略防止基于变压器的策略过拟合于视觉信息，并引导其正确关注力觉模态。实验表明，通过充分利用力觉信息，本方法在未见物体上的泛化性能相比无课程的基线方法提升了43%。视频演示、代码及使用说明详见https://jasonjzliu.com/factr/
原文摘要 · Abstract (English)
Many contact-rich tasks humans perform, such as box pickup or rolling dough, rely on force feedback for reliable execution. However, this force information, which is readily available in most robot arms, is not commonly used in teleoperation and policy learning. Consequently, robot behavior is often limited to quasi-static kinematic tasks that do not require intricate force-feedback. In this paper, we first present a low-cost, intuitive, bilateral teleoperation setup that relays external forces of the follower arm back to the teacher arm, facilitating data collection for complex, contact-rich tasks. We then introduce FACTR, a policy learning method that employs a curriculum which corrupts the visual input with decreasing intensity throughout training. The curriculum prevents our transformer-based policy from over-fitting to the visual input and guides the policy to properly attend to the force modality. We demonstrate that by fully utilizing the force information, our method significantly improves generalization to unseen objects by 43\% compared to baseline approaches without a curriculum. Video results, codebases, and instructions at https://jasonjzliu.com/factr/
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。