cVLA: Towards Efficient Camera-Space VLAs
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909971467206656 |
|---|---|
| author | Argus, Max Bratulic, Jelena Masnavi, Houman Velikanov, Maxim Heppert, Nick Valada, Abhinav Brox, Thomas |
| author_facet | Argus, Max Bratulic, Jelena Masnavi, Houman Velikanov, Maxim Heppert, Nick Valada, Abhinav Brox, Thomas |
| contents | Vision-Language-Action (VLA) models offer a compelling framework for tackling complex robotic manipulation tasks, but they are often expensive to train. In this paper, we propose a novel VLA approach that leverages the competitive performance of Vision Language Models (VLMs) on 2D images to directly infer robot end-effector poses in image frame coordinates. Unlike prior VLA models that output low-level controls, our model predicts trajectory waypoints, making it both more efficient to train and robot embodiment agnostic. Despite its lightweight design, our next-token prediction architecture effectively learns meaningful and executable robot trajectories. We further explore the underutilized potential of incorporating depth images, inference-time techniques such as decoding strategies, and demonstration-conditioned action generation. Our model is trained on a simulated dataset and exhibits strong sim-to-real transfer capabilities. We evaluate our approach using a combination of simulated and real data, demonstrating its effectiveness on a real robotic system. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_02190 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | cVLA: Towards Efficient Camera-Space VLAs Argus, Max Bratulic, Jelena Masnavi, Houman Velikanov, Maxim Heppert, Nick Valada, Abhinav Brox, Thomas Robotics Machine Learning Vision-Language-Action (VLA) models offer a compelling framework for tackling complex robotic manipulation tasks, but they are often expensive to train. In this paper, we propose a novel VLA approach that leverages the competitive performance of Vision Language Models (VLMs) on 2D images to directly infer robot end-effector poses in image frame coordinates. Unlike prior VLA models that output low-level controls, our model predicts trajectory waypoints, making it both more efficient to train and robot embodiment agnostic. Despite its lightweight design, our next-token prediction architecture effectively learns meaningful and executable robot trajectories. We further explore the underutilized potential of incorporating depth images, inference-time techniques such as decoding strategies, and demonstration-conditioned action generation. Our model is trained on a simulated dataset and exhibits strong sim-to-real transfer capabilities. We evaluate our approach using a combination of simulated and real data, demonstrating its effectiveness on a real robotic system. |
| title | cVLA: Towards Efficient Camera-Space VLAs |
| topic | Robotics Machine Learning |
| url | https://arxiv.org/abs/2507.02190 |