Multi-View Picking: Next-best-view Reaching for Improved Grasping in Clutter
This addresses the challenge of reliable robotic grasping in clutter, offering a practical improvement over static or fixed-view methods.
The paper tackled the problem of visual grasp detection in cluttered environments by developing a Multi-View Picking controller that selects informative viewpoints in real time to reduce uncertainty from occlusions, achieving 80% grasp success in trials, which is 12% higher than a single-viewpoint detector.
Camera viewpoint selection is an important aspect of visual grasp detection, especially in clutter where many occlusions are present. Where other approaches use a static camera position or fixed data collection routines, our Multi-View Picking (MVP) controller uses an active perception approach to choose informative viewpoints based directly on a distribution of grasp pose estimates in real time, reducing uncertainty in the grasp poses caused by clutter and occlusions. In trials of grasping 20 objects from clutter, our MVP controller achieves 80% grasp success, outperforming a single-viewpoint grasp detector by 12%. We also show that our approach is both more accurate and more efficient than approaches which consider multiple fixed viewpoints.