4 papers
Visual Manipulation with Legs
Xialin He, Chengjing Yuan, Wenxuan Zhou +3
Animals use limbs for both locomotion and manipulation. We aim to equip quadruped robots with similar versatility. This work introduces a system that enables quadruped robots to in…
HACMan: Learning Hybrid Actor-Critic Maps for 6D Non-Prehensile Manipulation
Wenxuan Zhou, Bowen Jiang, Fan Yang +2
Manipulating objects without grasping them is an essential component of human dexterity, referred to as non-prehensile manipulation. Non-prehensile manipulation may enable more com…
Reinforcement Learning in a Safety-Embedded MDP with Trajectory Optimization
Fan Yang, Wenxuan Zhou, Zuxin Liu +2
Safe Reinforcement Learning (RL) plays an important role in applying RL algorithms to safety-critical real-world applications, addressing the trade-off between maximizing rewards a…
HACMan++: Spatially-Grounded Motion Primitives for Manipulation
Bowen Jiang, Yilin Wu, Wenxuan Zhou +2
Although end-to-end robot learning has shown some success for robot manipulation, the learned policies are often not sufficiently robust to variations in object pose or geometry. T…