CAPGrasp: An $\mathbb{R}^3\times \text{SO(2)-equivariant}$ Continuous Approach-Constrained Generative Grasp Sampler
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909365828583424 |
|---|---|
| author | Weng, Zehang Lu, Haofei Lundell, Jens Kragic, Danica |
| author_facet | Weng, Zehang Lu, Haofei Lundell, Jens Kragic, Danica |
| contents | We propose CAPGrasp, an $\mathbb{R}^3\times \text{SO(2)-equivariant}$ 6-DoF continuous approach-constrained generative grasp sampler. It includes a novel learning strategy for training CAPGrasp that eliminates the need to curate massive conditionally labeled datasets and a constrained grasp refinement technique that improves grasp poses while respecting the grasp approach directional constraints. The experimental results demonstrate that CAPGrasp is more than three times as sample efficient as unconstrained grasp samplers while achieving up to 38% grasp success rate improvement. CAPGrasp also achieves 4-10% higher grasp success rates than constrained but noncontinuous grasp samplers. Overall, CAPGrasp is a sample-efficient solution when grasps must originate from specific directions, such as grasping in confined spaces. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2310_12113 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | CAPGrasp: An $\mathbb{R}^3\times \text{SO(2)-equivariant}$ Continuous Approach-Constrained Generative Grasp Sampler Weng, Zehang Lu, Haofei Lundell, Jens Kragic, Danica Robotics We propose CAPGrasp, an $\mathbb{R}^3\times \text{SO(2)-equivariant}$ 6-DoF continuous approach-constrained generative grasp sampler. It includes a novel learning strategy for training CAPGrasp that eliminates the need to curate massive conditionally labeled datasets and a constrained grasp refinement technique that improves grasp poses while respecting the grasp approach directional constraints. The experimental results demonstrate that CAPGrasp is more than three times as sample efficient as unconstrained grasp samplers while achieving up to 38% grasp success rate improvement. CAPGrasp also achieves 4-10% higher grasp success rates than constrained but noncontinuous grasp samplers. Overall, CAPGrasp is a sample-efficient solution when grasps must originate from specific directions, such as grasping in confined spaces. |
| title | CAPGrasp: An $\mathbb{R}^3\times \text{SO(2)-equivariant}$ Continuous Approach-Constrained Generative Grasp Sampler |
| topic | Robotics |
| url | https://arxiv.org/abs/2310.12113 |