Event
Successful real-world validation of whole-body control for a quadrupedal robotic arm on complex terrain
1 reports · 1 sources
Yanzhou Jin et al. proposed a reinforcement learning-based whole-body control framework for quadrupedal mobile manipulation and developed a lightweight quadrupedal robotic arm platform for real-world validation, successfully deploying it on slopes up to 30 degrees, steps, and deformable terrain. The framework combines hybrid task representation with task-priority whole-body learning to coordinate locomotion and manipulation while maintaining task consistency of end-effector behavior.
Report timeline
- Reinforcement Learning-Based Multi-Terrain Whole-Body Control for a Quadruped Mobile Manipulator
The paper proposes a reinforcement learning-based whole-body control framework for quadruped loco-manipulation on complex terrain. The framework combines a hybrid task representation with task-prioritized whole-body learning to coordinate locomotion and manipulation while preserving task-consistent end-effector behavior. The authors also developed a lightweight quadruped mobile manipulation platform for validation, and real-world experiments successfully deployed on complex terrains including slopes up to 30 degrees, stairs, and deformable terrain.
arXiv cs.RO ·