QuadWBG: Generalizable Quadrupedal Whole-Body Grasping
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915100325052416 |
|---|---|
| author | Wang, Jilong Rajabov, Javokhirbek Xu, Chaoyi Zheng, Yiming Wang, He |
| author_facet | Wang, Jilong Rajabov, Javokhirbek Xu, Chaoyi Zheng, Yiming Wang, He |
| contents | Legged robots with advanced manipulation capabilities have the potential to significantly improve household duties and urban maintenance. Despite considerable progress in developing robust locomotion and precise manipulation methods, seamlessly integrating these into cohesive whole-body control for real-world applications remains challenging. In this paper, we present a modular framework for robust and generalizable whole-body loco-manipulation controller based on a single arm-mounted camera. By using reinforcement learning (RL), we enable a robust low-level policy for command execution over 5 dimensions (5D) and a grasp-aware high-level policy guided by a novel metric, Generalized Oriented Reachability Map (GORM). The proposed system achieves state-of-the-art one-time grasping accuracy of 89% in the real world, including challenging tasks such as grasping transparent objects. Through extensive simulations and real-world experiments, we demonstrate that our system can effectively manage a large workspace, from floor level to above body height, and perform diverse whole-body loco-manipulation tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_06782 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | QuadWBG: Generalizable Quadrupedal Whole-Body Grasping Wang, Jilong Rajabov, Javokhirbek Xu, Chaoyi Zheng, Yiming Wang, He Robotics Artificial Intelligence Machine Learning Systems and Control Legged robots with advanced manipulation capabilities have the potential to significantly improve household duties and urban maintenance. Despite considerable progress in developing robust locomotion and precise manipulation methods, seamlessly integrating these into cohesive whole-body control for real-world applications remains challenging. In this paper, we present a modular framework for robust and generalizable whole-body loco-manipulation controller based on a single arm-mounted camera. By using reinforcement learning (RL), we enable a robust low-level policy for command execution over 5 dimensions (5D) and a grasp-aware high-level policy guided by a novel metric, Generalized Oriented Reachability Map (GORM). The proposed system achieves state-of-the-art one-time grasping accuracy of 89% in the real world, including challenging tasks such as grasping transparent objects. Through extensive simulations and real-world experiments, we demonstrate that our system can effectively manage a large workspace, from floor level to above body height, and perform diverse whole-body loco-manipulation tasks. |
| title | QuadWBG: Generalizable Quadrupedal Whole-Body Grasping |
| topic | Robotics Artificial Intelligence Machine Learning Systems and Control |
| url | https://arxiv.org/abs/2411.06782 |