VLSA: Vision-Language-Action Models with Plug-and-Play Safety Constraint Layer
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914199731437568 |
|---|---|
| author | Hu, Songqiao Liu, Zeyi Liu, Shuang Cen, Jun Meng, Zihan He, Xiao |
| author_facet | Hu, Songqiao Liu, Zeyi Liu, Shuang Cen, Jun Meng, Zihan He, Xiao |
| contents | Vision-Language-Action (VLA) models have demonstrated remarkable capabilities in generalizing across diverse robotic manipulation tasks. However, deploying these models in unstructured environments remains challenging due to the critical need for simultaneous task compliance and safety assurance, particularly in preventing potential collisions during physical interactions. In this work, we introduce a Vision-Language-Safe Action (VLSA) architecture, named AEGIS, which contains a plug-and-play safety constraint (SC) layer formulated via control barrier functions. AEGIS integrates directly with existing VLA models to improve safety with theoretical guarantees, while maintaining their original instruction-following performance. To evaluate the efficacy of our architecture, we construct a comprehensive safety-critical benchmark SafeLIBERO, spanning distinct manipulation scenarios characterized by varying degrees of spatial complexity and obstacle intervention. Extensive experiments demonstrate the superiority of our method over state-of-the-art baselines. Notably, AEGIS achieves a 59.16% improvement in obstacle avoidance rate while substantially increasing the task execution success rate by 17.25%. To facilitate reproducibility and future research, we make our code, models, and the benchmark datasets publicly available at https://vlsa-aegis.github.io/. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_11891 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | VLSA: Vision-Language-Action Models with Plug-and-Play Safety Constraint Layer Hu, Songqiao Liu, Zeyi Liu, Shuang Cen, Jun Meng, Zihan He, Xiao Robotics Systems and Control Vision-Language-Action (VLA) models have demonstrated remarkable capabilities in generalizing across diverse robotic manipulation tasks. However, deploying these models in unstructured environments remains challenging due to the critical need for simultaneous task compliance and safety assurance, particularly in preventing potential collisions during physical interactions. In this work, we introduce a Vision-Language-Safe Action (VLSA) architecture, named AEGIS, which contains a plug-and-play safety constraint (SC) layer formulated via control barrier functions. AEGIS integrates directly with existing VLA models to improve safety with theoretical guarantees, while maintaining their original instruction-following performance. To evaluate the efficacy of our architecture, we construct a comprehensive safety-critical benchmark SafeLIBERO, spanning distinct manipulation scenarios characterized by varying degrees of spatial complexity and obstacle intervention. Extensive experiments demonstrate the superiority of our method over state-of-the-art baselines. Notably, AEGIS achieves a 59.16% improvement in obstacle avoidance rate while substantially increasing the task execution success rate by 17.25%. To facilitate reproducibility and future research, we make our code, models, and the benchmark datasets publicly available at https://vlsa-aegis.github.io/. |
| title | VLSA: Vision-Language-Action Models with Plug-and-Play Safety Constraint Layer |
| topic | Robotics Systems and Control |
| url | https://arxiv.org/abs/2512.11891 |