arXiv：cs.LG· Hojoon Son, Fan Zhang·· 4 小时前AI 评分35
PG-VP：面向温度与辐射感知 VLA 导航的物理引导视觉提示
Seeing the Invisible: Physics-Guided Visual Prompting for Temperature- and Radiation-Aware VLA Navigation
AI 导读
研究人员提出物理引导视觉提示（PG-VP），一种即插即用多模态感知模块，复用冻结的 VLA 模型能力，通过叠加跨帧移动的虚拟障碍引导导航策略绕开不可见风险。在 OmniNav 的 R2R-CE 与 RxR-CE val-unseen 划分上，PG-VP 分别以 84.9% 和 83.2% 的比例引导策略采取低风险动作，导航成功率代价为 6.8 和 7.9 个百分点。
正文
Abstract:Vision-Language-Action (VLA) models have become a major paradigm for Vision-and-Language Navigation (VLN). However, in safety-critical facilities, invisible risks such as radiation or temperature spikes cannot be detected by an RGB camera, and handling each risk is expensive, requiring a new encoder, new data, and model retraining. We propose Physics-Guided Visual Prompting (PG-VP), a plug-and-play multimodal perception module that instead reuses what a frozen VLA model already does well: avoiding visible obstacles. Given a proximal radiation or thermal source, PG-VP performs a physics-guided risk assessment to determine the avoidance direction and overlays a corresponding virtual obstacle that moves across consecutive frames (Dynamic Visual Prompting). The navigation policy then naturally detours around this invisible hazard. The identical virtual obstacle is used regardless of hazard type, so the visual prompting pattern remains fixed as sensors are added. When no hazard is detected, nothing is rendered, and the policy behaves exactly as it would without PG-VP. We evaluate PG-VP on OmniNav using the val-unseen splits of R2R-CE and RxR-CE, where it guides the policy toward intended low-risk actions in 84.9% and 83.2% of cases, at a cost of 6.8 and 7.9 percentage points in navigation success rate. We further test it with distinct scenarios on a real robot in the presence of actual thermal and radiation sources, all without any retraining. The real test shows that PG-VP effectively avoids these invisible hazards, improving worst-10% average trajectory safety by 63.45% and 32.59% against thermal and radiation sources, respectively.
| Comments: | 8 pages, 7 figures, 2 tables |
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.07558 [cs.RO] |
| (or arXiv:2610.07558v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.07558 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Hojoon Son [view email]
[v1]
Tue, 6 Oct 2026 00:44:06 UTC (3,560 KB)
来源：arXiv：cs.LG · arxiv.org