Safe-Night VLA: Seeing the Unseen via Thermal-Perceptive Vision-Language-Action Models for Safety-Critical Manipulation
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911490321154048 |
|---|---|
| author | Yu, Dian Zhou, Qingchuan Huang, Bingkun Khadiv, Majid Yang, Zewen |
| author_facet | Yu, Dian Zhou, Qingchuan Huang, Bingkun Khadiv, Majid Yang, Zewen |
| contents | Current Vision-Language-Action (VLA) models rely primarily on RGB perception, preventing them from capturing modalities such as thermal signals that are imperceptible to conventional visual sensors. Moreover, end-to-end generative policies lack explicit safety constraints, making them fragile when encountering obstacles and novel scenarios outside the training distribution. To address these limitations, we propose Safe-Night VLA, a multimodal manipulation framework that enables robots to see the unseen while enforcing rigorous safety constraints for thermal-aware manipulation in unstructured environments. Specifically, Safe-Night VLA integrates long-wave infrared thermal perception into a pre-trained vision-language backbone, enabling semantic reasoning grounded in thermodynamic properties. To ensure safe execution under out-of-distribution conditions, we incorporate a safety filter via control barrier functions, which provide deterministic workspace constraint enforcement during policy execution. We validate our framework through real-world experiments on a Franka manipulator, introducing a novel evaluation paradigm featuring temperature-conditioned manipulation, subsurface target localization, and reflection disambiguation, while maintaining constrained execution at inference time. Results demonstrate that Safe-Night VLA outperforms RGB-only baselines and provide empirical evidence that foundation models can effectively leverage non-visible physical modalities for robust manipulation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_05754 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Safe-Night VLA: Seeing the Unseen via Thermal-Perceptive Vision-Language-Action Models for Safety-Critical Manipulation Yu, Dian Zhou, Qingchuan Huang, Bingkun Khadiv, Majid Yang, Zewen Robotics Current Vision-Language-Action (VLA) models rely primarily on RGB perception, preventing them from capturing modalities such as thermal signals that are imperceptible to conventional visual sensors. Moreover, end-to-end generative policies lack explicit safety constraints, making them fragile when encountering obstacles and novel scenarios outside the training distribution. To address these limitations, we propose Safe-Night VLA, a multimodal manipulation framework that enables robots to see the unseen while enforcing rigorous safety constraints for thermal-aware manipulation in unstructured environments. Specifically, Safe-Night VLA integrates long-wave infrared thermal perception into a pre-trained vision-language backbone, enabling semantic reasoning grounded in thermodynamic properties. To ensure safe execution under out-of-distribution conditions, we incorporate a safety filter via control barrier functions, which provide deterministic workspace constraint enforcement during policy execution. We validate our framework through real-world experiments on a Franka manipulator, introducing a novel evaluation paradigm featuring temperature-conditioned manipulation, subsurface target localization, and reflection disambiguation, while maintaining constrained execution at inference time. Results demonstrate that Safe-Night VLA outperforms RGB-only baselines and provide empirical evidence that foundation models can effectively leverage non-visible physical modalities for robust manipulation. |
| title | Safe-Night VLA: Seeing the Unseen via Thermal-Perceptive Vision-Language-Action Models for Safety-Critical Manipulation |
| topic | Robotics |
| url | https://arxiv.org/abs/2603.05754 |