WBAG：面向视觉-语言-动作操控的全身体与附着几何安全框架

中文摘要 AI 辅助

视觉-语言-动作（VLA）策略在泛化机器人操控方面展现出令人印象深刻的能力，但其在现实世界中的部署仍面临挑战，原因在于机器人不同部件、被操控物体以及周围环境之间可能发生的碰撞。现有的推理时VLA安全框架通常依赖于简化的以末端执行器为中心的表示，这些表示并未显式建模完整的铰接式机器人及附着物体的几何形状。在本文中，我们提出了WBAG，一种对机器人全身体及与抓取相关的附着几何进行建模的安全框架。WBAG构建了一个抓取条件化的安全集，该安全集在物体被抓取时调整受保护的几何形状，然后将这一动态演化的几何形状转换为可微分的控制屏障函数（CBF）约束，以最小程度地修改VLA原生的六维操作空间动作，从而在机器人、场景和附着几何之间实现碰撞避免。在SafeLIBERO基准（一个为安全评估而添加了障碍物的LIBERO变体）上，在监控所有符合条件的非任务物体的场景级安全评估器下，WBAG在评估的方法中取得了最佳的整体安全性和安全任务成功率，达到了97.38%的聚合场景安全性和59.38%的安全成功率。

英文摘要

Vision-language-action (VLA) policies have demonstrated impressive capabilities in generalizable robotic manipulation, but their deployment in the real world remains challenging due to potential collisions involving different parts of the robot, manipulated objects, and the surrounding environment. Existing inference-time VLA safety frameworks typically rely on simplified end-effector-centered representations that do not explicitly model the full articulated robot and attached-object geometry. In this paper, we present WBAG, a safety framework that models the robot's whole-body and grasp-dependent attached geometry. WBAG constructs a grasp-conditioned safe set that adapts the protected geometry as objects are grasped, then converts this evolving geometry into differentiable CBF constraints that minimally modify the VLA's six-dimensional operational-space action for collision avoidance across robot, scene, and attached geometry. On the SafeLIBERO benchmark, a variant of LIBERO augmented with obstacles for safety evaluation, WBAG achieves the best overall safety and safe task success among the evaluated methods under a scene-level safety evaluator that monitors all eligible non-task objects, reaching 97.38% aggregate Scene Safety and 59.38% Safe Success. Project page: https://samuelzhen.com/projects/wbag

发表机构

• Texas A&M University（德克萨斯A&M大学）
• University of Pennsylvania（宾夕法尼亚大学）
• University of Illinois Chicago（伊利诺伊大学芝加哥分校）

机构由 AI 辅助整理，请以论文原文为准。

补充信息

备注

Project page: https://samuelzhen.com/projects/wbag