Learning Continuous 3D Reconstructions for Geometrically Aware Grasping
This addresses the challenge of geometric awareness in robotic grasping for improved reliability, though it is incremental by building on existing deep learning approaches.
The paper tackles the problem of enabling robots to explicitly reason about full 3D geometry for grasping unseen objects from partial views, resulting in improved grasp synthesis through a learned 3D reconstruction and continuous optimization, with evaluation in 96 robot trials.
Deep learning has enabled remarkable improvements in grasp synthesis for previously unseen objects from partial object views. However, existing approaches lack the ability to explicitly reason about the full 3D geometry of the object when selecting a grasp, relying on indirect geometric reasoning derived when learning grasp success networks. This abandons explicit geometric reasoning, such as avoiding undesired robot object collisions. We propose to utilize a novel, learned 3D reconstruction to enable geometric awareness in a grasping system. We leverage the structure of the reconstruction network to learn a grasp success classifier which serves as the objective function for a continuous grasp optimization. We additionally explicitly constrain the optimization to avoid undesired contact, directly using the reconstruction. We examine the role of geometry in grasping both in the training of grasp metrics and through 96 robot grasping trials. Our results can be found on https://sites.google.com/view/reconstruction-grasp/.