6-DoF Contrastive Grasp Proposal Network

TL;DR

This paper introduces CGPN, a neural network that predicts 6-DoF grasp poses from a single depth image, using contrastive learning to enhance real-world robustness and enabling fast, collision-free grasp detection for robotic manipulation.

Contribution

The paper presents a novel 6-DoF grasp proposal network that incorporates contrastive learning and synthetic training to improve grasp prediction accuracy and robustness from single-view depth images.

Findings

Locates collision-free grasps within 0.5 seconds

Effective in real-world robotic experiments

Bridges simulation-to-real gap with contrastive learning

Abstract

Proposing grasp poses for novel objects is an essential component for any robot manipulation task. Planning six degrees of freedom (DoF) grasps with a single camera, however, is challenging due to the complex object shape, incomplete object information, and sensor noise. In this paper, we present a 6-DoF contrastive grasp proposal network (CGPN) to infer 6-DoF grasps from a single-view depth image. First, an image encoder is used to extract the feature map from the input depth image, after which 3-DoF grasp regions are proposed from the feature map with a rotated region proposal network. Feature vectors that within the proposed grasp regions are then extracted and refined to 6-DoF grasps. The proposed model is trained offline with synthetic grasp data. To improve the robustness in reality and bridge the simulation-to-real gap, we further introduce a contrastive learning module and variant image processing techniques during the training. CGPN can locate collision-free grasps of an object using a single-view depth image within 0.5 seconds. Experiments on a physical robot further demonstrate the effectiveness of the algorithm.