Grasp Proposal Networks: An End-to-End Solution for Visual Learning of Robotic Grasps

TL;DR

This paper introduces GPNet, an end-to-end neural network that predicts diverse 6-DOF robotic grasps from a single view, improving efficiency and coverage over heuristic sampling methods.

Contribution

The paper proposes a novel grasp proposal network with anchor-based grasp centers, enabling diverse and efficient grasp predictions for unseen objects from limited views.

Findings

GPNet outperforms existing methods in simulation tests.

The dataset and approach facilitate better real-world grasping.

Enhanced coverage improves grasp success rates.

Abstract

Learning robotic grasps from visual observations is a promising yet challenging task. Recent research shows its great potential by preparing and learning from large-scale synthetic datasets. For the popular, 6 degree-of-freedom (6-DOF) grasp setting of parallel-jaw gripper, most of existing methods take the strategy of heuristically sampling grasp candidates and then evaluating them using learned scoring functions. This strategy is limited in terms of the conflict between sampling efficiency and coverage of optimal grasps. To this end, we propose in this work a novel, end-to-end \emph{Grasp Proposal Network (GPNet)}, to predict a diverse set of 6-DOF grasps for an unseen object observed from a single and unknown camera view. GPNet builds on a key design of grasp proposal module that defines \emph{anchors of grasp centers} at discrete but regular 3D grid corners, which is flexible to support either more precise or more diverse grasp predictions. To test GPNet, we contribute a synthetic dataset of 6-DOF object grasps; evaluation is conducted using rule-based criteria, simulation test, and real test. Comparative results show the advantage of our methods over existing ones. Notably, GPNet gains better simulation results via the specified coverage, which helps achieve a ready translation in real test. We will make our dataset publicly available.