arXiv cs.AI· Haonan Chen, Feiyang Wu, Yuxiang Ma, Mustafa Mete, Pengfei Ye, Junxuan Shen, Cheng Zhu, Aurora Ruggeri, Kelvin Cheung, Jiayuan Mao, Edward Adelson, Jiajun Wu, Robert D. Howe, Yilun Du·· 13 小时前AI 评分35
利用视觉预测力实现机器人学习
Robot Learning with Visual Predicted Force
AI 导读
研究提出通过 Fin Ray 柔性夹爪的形变进行视觉力预测，无需专用力或触觉传感器即可实现力感知操作。方法先训练视觉力估计器为演示数据标注力估计，再训练动作-力提议策略联合生成候选动作及其预期力，测试时执行预测力最接近目标值的动作。在浆果采摘、空罐抓取、手内重定向和插头插入任务上验证了该方法可指导推理阶段的动作选择。
正文 · 原文
Authors:Haonan Chen, Feiyang Wu, Yuxiang Ma, Mustafa Mete, Pengfei Ye, Junxuan Shen, Cheng Zhu, Aurora Ruggeri, Kelvin Cheung, Jiayuan Mao, Edward Adelson, Jiajun Wu, Robert D. Howe, Yilun Du
Abstract:Force-aware manipulation typically relies on specialized force or tactile sensors. We show that force-aware manipulation can instead be achieved through visual force prediction from the deformation of a compliant Fin Ray gripper. Our approach trains two models. First, we train a visual force estimator on calibration data and use it to annotate task demonstrations with force estimates. Second, we train an action--force proposal policy on these force-augmented demonstrations to jointly generate candidate robot actions and their associated forces. At test time, we sample candidate actions and the forces they are expected to produce, then execute the action whose predicted force is closest to a target from the demonstrations. We evaluate our approach on berry picking, empty-can grasping, in-hand reorientation, and plug insertion. Our results show that visual force prediction can guide inference-time action selection for contact-rich manipulation without requiring force or tactile sensors at deployment.
| Comments: | 9 pages, 8 figures. Project Website: this https URL |
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.04741 [cs.RO] |
| (or arXiv:2610.04741v2 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.04741 arXiv-issued DOI via DataCite |
Submission history
From: Haonan Chen [view email]
[v1]
Sat, 3 Oct 2026 20:11:33 UTC (9,579 KB)
[v2]
Tue, 20 Oct 2026 19:20:24 UTC (9,553 KB)
来源：arXiv cs.AI · arxiv.org