Spatial RoboGrasp: Generalized Robotic Grasping Control Policy
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909624850972672 |
|---|---|
| author | Huang, Yiqi Davies, Travis Yan, Jiahuan Sun, Jiankai Chen, Xiang Hu, Luhui |
| author_facet | Huang, Yiqi Davies, Travis Yan, Jiahuan Sun, Jiankai Chen, Xiang Hu, Luhui |
| contents | Achieving generalizable and precise robotic manipulation across diverse environments remains a critical challenge, largely due to limitations in spatial perception. While prior imitation-learning approaches have made progress, their reliance on raw RGB inputs and handcrafted features often leads to overfitting and poor 3D reasoning under varied lighting, occlusion, and object conditions. In this paper, we propose a unified framework that couples robust multimodal perception with reliable grasp prediction. Our architecture fuses domain-randomized augmentation, monocular depth estimation, and a depth-aware 6-DoF Grasp Prompt into a single spatial representation for downstream action planning. Conditioned on this encoding and a high-level task prompt, our diffusion-based policy yields precise action sequences, achieving up to 40% improvement in grasp success and 45% higher task success rates under environmental variation. These results demonstrate that spatially grounded perception, paired with diffusion-based imitation learning, offers a scalable and robust solution for general-purpose robotic grasping. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_20814 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Spatial RoboGrasp: Generalized Robotic Grasping Control Policy Huang, Yiqi Davies, Travis Yan, Jiahuan Sun, Jiankai Chen, Xiang Hu, Luhui Robotics Computer Vision and Pattern Recognition Achieving generalizable and precise robotic manipulation across diverse environments remains a critical challenge, largely due to limitations in spatial perception. While prior imitation-learning approaches have made progress, their reliance on raw RGB inputs and handcrafted features often leads to overfitting and poor 3D reasoning under varied lighting, occlusion, and object conditions. In this paper, we propose a unified framework that couples robust multimodal perception with reliable grasp prediction. Our architecture fuses domain-randomized augmentation, monocular depth estimation, and a depth-aware 6-DoF Grasp Prompt into a single spatial representation for downstream action planning. Conditioned on this encoding and a high-level task prompt, our diffusion-based policy yields precise action sequences, achieving up to 40% improvement in grasp success and 45% higher task success rates under environmental variation. These results demonstrate that spatially grounded perception, paired with diffusion-based imitation learning, offers a scalable and robust solution for general-purpose robotic grasping. |
| title | Spatial RoboGrasp: Generalized Robotic Grasping Control Policy |
| topic | Robotics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.20814 |