Guardado en:
| Autores principales: | , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2603.22899 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908909347799040 |
|---|---|
| author | Yan, Teng Pei, Zhengyang Shi, Chengyu Yu, Yue Chen, Yikun Zhu, Zilong Fang, Zelin Guo, Kaile Wang, Zihang Tian, Peigen Zhong, Bingzhuo |
| author_facet | Yan, Teng Pei, Zhengyang Shi, Chengyu Yu, Yue Chen, Yikun Zhu, Zilong Fang, Zelin Guo, Kaile Wang, Zihang Tian, Peigen Zhong, Bingzhuo |
| contents | Deploying Vision-Language-Action (VLA) models on resource-constrained edge platforms encounters a fundamental conflict between high-latency semantic inference and the high-frequency control required for dynamic manipulation. To address the challenge, this paper presents Agile-VLA, a hierarchical framework designed for industrial pose reorientation tasks on edge devices such as the NVIDIA Jetson Orin Nano. The core innovation is an Implicit Affordance Anchoring mechanism that directly maps geometric visual cues, specifically centroid and rim keypoint anchors, into structured parametric action primitives, thereby substantially reducing reliance on high-latency semantic inference during closed-loop control. By decoupling perception (10 Hz) from control (50 Hz) via an asynchronous dual-stream architecture, the system effectively mitigates the frequency mismatch inherent in edge-based robot learning. Experimental results on a standard 6-DoF manipulator demonstrate that Agile-VLA achieves robust rectification of complex, irregular workpieces using only 5-shot demonstrations through extrinsic dexterity. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_22899 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Agile-VLA: Few-Shot Industrial Pose Rectification via Implicit Affordance Anchoring Yan, Teng Pei, Zhengyang Shi, Chengyu Yu, Yue Chen, Yikun Zhu, Zilong Fang, Zelin Guo, Kaile Wang, Zihang Tian, Peigen Zhong, Bingzhuo Robotics Deploying Vision-Language-Action (VLA) models on resource-constrained edge platforms encounters a fundamental conflict between high-latency semantic inference and the high-frequency control required for dynamic manipulation. To address the challenge, this paper presents Agile-VLA, a hierarchical framework designed for industrial pose reorientation tasks on edge devices such as the NVIDIA Jetson Orin Nano. The core innovation is an Implicit Affordance Anchoring mechanism that directly maps geometric visual cues, specifically centroid and rim keypoint anchors, into structured parametric action primitives, thereby substantially reducing reliance on high-latency semantic inference during closed-loop control. By decoupling perception (10 Hz) from control (50 Hz) via an asynchronous dual-stream architecture, the system effectively mitigates the frequency mismatch inherent in edge-based robot learning. Experimental results on a standard 6-DoF manipulator demonstrate that Agile-VLA achieves robust rectification of complex, irregular workpieces using only 5-shot demonstrations through extrinsic dexterity. |
| title | Agile-VLA: Few-Shot Industrial Pose Rectification via Implicit Affordance Anchoring |
| topic | Robotics |
| url | https://arxiv.org/abs/2603.22899 |