Event
Authors Propose Visibility-Aware Mobile Grasping System for Dynamic Environments
1 reports · 1 sources
Tianrun Hu and other authors propose a unified system that simultaneously constrains the collision risk of whole-body trajectories and the target visibility during grasping in unknown and dynamic environments. The system consists of three parts: a velocity-aware gaze policy, a hierarchical subgoal policy, and a receding-horizon whole-body planner. In 400 randomized simulation scenarios and real-world trials with a Fetch mobile manipulation robot, the system achieved success rates of 70.4% in unknown static environments and 57.9% in dynamic environments, outperforming decoupled navigation and manipulation policies by 16.3 and 9.3 percentage points, respectively, and significantly reducing collisions. These results correspond to the setting using a limited field-of-view camera.
Report timeline
- Visibility-aware mobile grasping system for dynamic environments
The authors propose a unified system that simultaneously constrains collision risk along the whole-body trajectory and target visibility during grasping in unknown and dynamic environments, comprising three parts: a velocity-aware gaze policy, a hierarchical subgoal policy, and a receding-horizon whole-body planner. In 400 randomized simulation scenarios and real-world trials on a Fetch mobile manipulator, the system achieved success rates of 70.4% in unknown static environments and 57.9% in dynamic environments, 16.3 and 9.3 percentage points higher than decoupled navigation and manipulation policies, respectively, while significantly reducing collisions.
arXiv cs.RO ·