arXiv：cs.AI· Weikang Wan, Zeyu Lai, Jianyu Hou, Yuchen Zhou, Jiawei Fu, Henrik I Christensen, Hao Su·· 19 小时前
DexFLEX：面向指令引导灵巧操作的接触感知基础控制器
DexFLEX: Contact-Aware Foundation Controller for Command-Guided Dexterity
AI 导读
DexFLEX 是一个接触感知基础控制器，将指尖运动草案转为接触一致的关节指令，通过 Draft-Dream-Select 推理流程在动作块中预测接触后果并择优执行。同一训练完成的控制器无需重训即可提升鲁棒性，真实世界遥操作成功率从 29.2% 提升至 78.3%。在策略学习中成功率达 46.7%，为直接关节动作预测的 3.5 倍、仅先验纠错的 1.56 倍。
正文
Abstract:Dexterous robot hands can receive useful motion intent from teleoperation or learned policies, but successful execution also depends on contact decisions that these commands rarely specify. We introduce DexFLEX, a contact-aware foundation controller that turns upstream fingertip-motion drafts into contact-consistent joint commands. Instead of treating a draft as a trajectory to copy, DexFLEX treats it as evidence about intent: from the current tactile-proprioceptive state, it proposes feasible short-horizon motion chunks, predicts their contact consequences, and selects the candidate that best follows the command while preserving future contact stability. At inference time, a simple Draft-Dream-Select procedure combines pure-prior proposals for recovery with draft-seeded proposals for responsiveness, then decodes the selected motion into executable joint-space targets. Across degraded-command simulation, real-world shared control, and visuomotor policy learning, the same trained controller improves robustness without retraining, increasing real-world teleoperation success from 29.2% to 78.3% and reaching 46.7% policy-learning success, 3.5 times direct joint-action prediction, and 1.56 times prior-only correction. Videos are available at this http URL.
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.14372 [cs.RO] |
| (or arXiv:2610.14372v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.14372 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Weikang Wan [view email]
[v1]
Fri, 9 Oct 2026 22:47:28 UTC (14,420 KB)
来源：arXiv：cs.AI · arxiv.org