arXiv 2603.152572026-08-04cs.RO

HapticVLA：通过视觉-语言-动作模型实现接触丰富的操作，无需推理时触觉感知

HapticVLA: Contact-Rich Manipulation via Vision-Language-Action Model without Inference-Time Tactile Sensing

Intelligent Space Robotics Laboratory, Skolkovo Institute of Science and Technology（智能空间机器人实验室，斯克洛夫科学与技术研究所）

Konstantin Gubernatorov, Mikhail Sannikov, Ilya Mikhalchuk, Egor Kuznetsov, Makar Artemov, Ogunwoye Faith Oluwatobi, Marcelino Fernando, Artem Asanov, Ziang Guo… 展开作者

Konstantin Gubernatorov, Mikhail Sannikov, Ilya Mikhalchuk, Egor Kuznetsov, Makar Artemov, Ogunwoye Faith Oluwatobi, Marcelino Fernando, Artem Asanov, Ziang Guo, Dzmitry Tsetserukou

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

HapticVLA通过视觉-语言-动作模型实现接触丰富的操作，无需触觉传感器，在现实实验中成功率达到86.7%。

摘要：

触觉感知是视觉-语言-动作（VLA）架构的关键能力，因为它能够实现接触丰富的任务中灵巧且安全的操作。然而，依赖专用触觉硬件会增加成本并降低在不同机器人平台上的可重复性。我们主张触觉感知的操作可以在离线学习后部署，而无需直接的触觉反馈。为此，我们提出了HapticVLA，其分为两个紧密耦合的阶段：安全感知奖励加权流匹配（SA-RWFM）和触觉蒸馏（TD）。SA-RWFM训练了一个流匹配的动作专家，该专家结合了预先计算的安全感知触觉奖励，惩罚过度的抓取力和不优的抓取轨迹。TD进一步将这种触觉感知能力转移到传统的VLA：我们从SA-RWFM教师中蒸馏出一个紧凑的触觉标记，并训练一个学生VLA来从视觉和状态模态预测该标记，从而在推理时无需使用板载触觉传感器即可实现触觉感知的动作生成。这种设计在VLA中保留了接触丰富的触觉感知推理，同时在部署时不需要板载触觉传感器。在现实实验中，HapticVLA实现了86.7%的平均成功率，持续优于基线VLA——包括在推理时提供直接触觉反馈的版本。

英文摘要：

Tactile sensing is a crucial capability for Vision-Language-Action (VLA) architectures, as it enables dexterous and safe manipulation in contact-rich tasks. However, reliance on dedicated tactile hardware increases cost and reduces reproducibility across robotic platforms. We argue that tactile-aware manipulation can be learned offline and deployed without direct haptic feedback at inference. To this end, we present HapticVLA, which proceeds in two tightly coupled stages: Safety-Aware Reward-Weighted Flow Matching (SA-RWFM) and Tactile Distillation (TD). SA-RWFM trains a flow-matching action expert that incorporates precomputed, safety-aware tactile rewards penalizing excessive grasping force and suboptimal grasping trajectories. TD further transfers this tactile-aware capability into a conventional VLA: we distill a compact tactile token from the SA-RWFM teacher and train a student VLA to predict that token from vision and state modalities, enabling tactile-aware action generation at inference without requiring on-board tactile sensors. This design preserves contact-rich tactile-aware reasoning within VLA while removing the need for on-board tactile sensors during deployment. On real-world experiments, HapticVLA achieves a mean success rate of 86.7%, consistently outperforming baseline VLAs - including versions provided with direct tactile feedback during inference.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。