Skill-Nav: Enhanced Navigation with Versatile Quadrupedal Locomotion via Waypoint Interface
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911139939483648 |
|---|---|
| author | Wang, Dewei Bai, Chenjia Li, Chenhui Shi, Jiyuan Ding, Yan Zhang, Chi Zhao, Bin |
| author_facet | Wang, Dewei Bai, Chenjia Li, Chenhui Shi, Jiyuan Ding, Yan Zhang, Chi Zhao, Bin |
| contents | Quadrupedal robots have demonstrated exceptional locomotion capabilities through Reinforcement Learning (RL), including extreme parkour maneuvers. However, integrating locomotion skills with navigation in quadrupedal robots has not been fully investigated, which holds promise for enhancing long-distance movement capabilities. In this paper, we propose Skill-Nav, a method that incorporates quadrupedal locomotion skills into a hierarchical navigation framework using waypoints as an interface. Specifically, we train a waypoint-guided locomotion policy using deep RL, enabling the robot to autonomously adjust its locomotion skills to reach targeted positions while avoiding obstacles. Compared with direct velocity commands, waypoints offer a simpler yet more flexible interface for high-level planning and low-level control. Utilizing waypoints as the interface allows for the application of various general planning tools, such as large language models (LLMs) and path planning algorithms, to guide our locomotion policy in traversing terrains with diverse obstacles. Extensive experiments conducted in both simulated and real-world scenarios demonstrate that Skill-Nav can effectively traverse complex terrains and complete challenging navigation tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_21853 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Skill-Nav: Enhanced Navigation with Versatile Quadrupedal Locomotion via Waypoint Interface Wang, Dewei Bai, Chenjia Li, Chenhui Shi, Jiyuan Ding, Yan Zhang, Chi Zhao, Bin Robotics Quadrupedal robots have demonstrated exceptional locomotion capabilities through Reinforcement Learning (RL), including extreme parkour maneuvers. However, integrating locomotion skills with navigation in quadrupedal robots has not been fully investigated, which holds promise for enhancing long-distance movement capabilities. In this paper, we propose Skill-Nav, a method that incorporates quadrupedal locomotion skills into a hierarchical navigation framework using waypoints as an interface. Specifically, we train a waypoint-guided locomotion policy using deep RL, enabling the robot to autonomously adjust its locomotion skills to reach targeted positions while avoiding obstacles. Compared with direct velocity commands, waypoints offer a simpler yet more flexible interface for high-level planning and low-level control. Utilizing waypoints as the interface allows for the application of various general planning tools, such as large language models (LLMs) and path planning algorithms, to guide our locomotion policy in traversing terrains with diverse obstacles. Extensive experiments conducted in both simulated and real-world scenarios demonstrate that Skill-Nav can effectively traverse complex terrains and complete challenging navigation tasks. |
| title | Skill-Nav: Enhanced Navigation with Versatile Quadrupedal Locomotion via Waypoint Interface |
| topic | Robotics |
| url | https://arxiv.org/abs/2506.21853 |