从二值可行性 Oracle 学习安全的 Flow Matching 机器人策略
Learning Safe Flow Matching Robot Policies from Binary Feasibility Oracles
AI 导读
研究者提出一种 oracle 引导的可微安全防护方法，用于 flow matching 机器人策略在仅有黑盒二值可行性查询时满足物理约束。该方法在前向传播中纠正不安全动作更新，并从局部边界信息提取约束感知学习信号，训练与推理共用同一防护机制。在障碍感知机械臂操作实验中实现 100% 约束满足率，同时保持任务性能。
正文 · 原文
Abstract:Robot foundation models increasingly rely on flow-based action generation for general-purpose manipulation, yet the generated actions are not guaranteed to satisfy physical constraints such as joint limits and collision avoidance. This problem is especially challenging when the feasible set cannot be described analytically and is accessible only through black-box feasibility oracles. We study this practical oracle-only setting and propose an oracle-guided differentiable safeguard for safe flow-matching robot policies. The safeguard uses binary feasibility queries to correct unsafe updates in the forward pass, while extracting local boundary information to provide constraint-aware learning signals in the backward pass. Combined with discrete-time flow matching, the same safeguard is used during training and inference, enabling forward--backward aligned policy learning without explicit or differentiable constraint models. Experiments on obstacle-aware robotic manipulation demonstrate 100% constraint satisfaction while preserving task performance.
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.40075 [cs.RO] |
| (or arXiv:2610.40075v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.40075 arXiv-issued DOI via DataCite |
Submission history
From: Jianming Ma [view email]
[v1]
Thu, 29 Oct 2026 13:53:04 UTC (465 KB)
来源：arXiv cs.AI · arxiv.org