用集合传播方法提升视觉运动策略的可达动作验证精度
Probabilistic Reachable-Action Verification of Visuomotor Policies via Set-Based Training
- 冻结视觉编码器，只在低维接口做集合传播以降低计算成本
- 通过区间传播优化终端动作包围盒宽度，实测可达半径缩小30%以上
- 适合对机器人动作可靠性有严格要求的闭环控制场景
视觉运动策略的可达性分析因大型视觉编码器导致端到端集合传播计算昂贵且过于保守。为此，本文冻结视觉编码器，仅在编码器与下游策略间的低维接口进行集合传播，并利用预留的相机位姿扰动校准接口集合。通过使用区间（zonotopes）传播该集合，直接优化终端输出包围盒宽度。评估时，从预设分布采样相机位姿扰动，利用滚动级分拆置信校准将动作偏差得分转化为具有有限样本覆盖率的概率可达动作半径。在受控操作实验中，基于集合的训练显著缩小了该半径，同时保持闭环任务能力；而行为仅限、观测一致性及点式对抗等对照方法均产生更大的半径。
原文摘要 · Abstract (English)
Reachability analysis for visuomotor policies is difficult because large visual encoders make end-to-end set propagation computationally expensive and excessively conservative. We therefore freeze the visual encoder and confine set propagation to a low-dimensional interface between it and the downstream policy, with the interface set calibrated from held-out camera-pose perturbations. Propagating this set through the policy with zonotopes yields a terminal output-enclosure width that set-based training optimizes directly. During evaluation, camera-pose perturbations are sampled from the prescribed distribution, and rollout-level split conformal calibration converts the resulting action-deviation scores into a probabilistic reachable-action radius with finite-sample coverage. In controlled manipulation experiments, set-based training reduces this radius while preserving closed-loop task capability, and matched behavior-only, observational-consistency, and pointwise-adversarial controls all leave a larger radius.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。