VLPSA：用于学习策略全身安全的视觉-语言-泊松安全动作

中文摘要 AI 辅助

视觉-语言-动作（VLA）模型使得机器人操作越来越通用化，但此类学习策略无法提供碰撞避免的安全保证，尤其是在训练分布之外的环境中。本工作提出了视觉-语言-泊松安全动作（VLPSA），一种安全过滤框架，无需重新训练即可在杂乱和动态环境中为VLA策略提供全身安全。VLPSA从感知数据在线合成泊松安全函数（PSF），生成控制屏障函数（CBF），并通过CBF-QP安全过滤器在整个身体以及任何被抓取物体（视为最终机器人连杆的延伸）上强制执行。为了在保持关键任务区域精细空间分辨率的同时实现实时部署，VLPSA使用布尔CBF组合结合了该PSF的双分辨率。我们在SafeLIBERO上对VLPSA与安全过滤基线进行了评估，其在所评估方法中实现了最高的碰撞避免率，将基础π0.5策略的碰撞避免率从23.1%提高到91.2%，同时超越了其任务成功率。我们进一步在Franka FR3上部署了VLPSA，在具有动态障碍物和人为干扰的杂乱场景中，展示了操作任务期间的实时全身安全。

英文摘要

Vision-language-action (VLA) models enable increasingly general-purpose robotic manipulation, but such learned policies do not provide safety guarantees for collision avoidance---especially in environments outside of training distributions. This work presents Vision-Language-Poisson-Safe Actions (VLPSA), a safety filtering framework that provides full-body safety for VLA policies in cluttered and dynamic environments without retraining. VLPSA synthesizes Poisson Safety Functions (PSF) online from perception data, yielding a Control Barrier Function (CBF) that is enforced through a CBF-QP safety filter over the full body and any grasped object, treated as an extension of the final robot link. To enable real-time deployment while maintaining fine spatial resolution in critical task regions, VLPSA combines dual resolutions of this PSF using Boolean CBF compositions. We evaluate VLPSA on SafeLIBERO against safety-filtering baselines, where it achieves the highest collision avoidance rate among the evaluated methods, increasing collision avoidance from 23.1% for the base $π_{0.5}$ policy to 91.2% while surpassing its task success rate. We further deploy VLPSA on a Franka FR3 in cluttered scenes with dynamic obstacles and human interference, demonstrating real-time full-body safety during manipulation tasks.

发表机构

• California Institute of Technology（加州理工学院）

机构由 AI 辅助整理，请以论文原文为准。