GP-net: Flexible Viewpoint Grasp Proposal

TL;DR

GP-net is a CNN model that generates 6-DoF grasps from flexible viewpoints, trained on synthetic data, and outperforms existing methods in real-world mobile manipulation tasks.

Contribution

Introduces GP-net, a novel CNN for flexible viewpoint grasping, trained on synthetic data, and validated with real-world experiments showing improved success rates.

Findings

GP-net achieves 54.4% grasp success rate.

Outperforms VGN and GPD in real-world tests.

Supports flexible, unknown viewpoints without workspace constraints.

Abstract

We present the Grasp Proposal Network (GP-net), a Convolutional Neural Network model which can generate 6-DoF grasps from flexible viewpoints, e.g. as experienced by mobile manipulators. To train GP-net, we synthetically generate a dataset containing depth-images and ground-truth grasp information. In real-world experiments, we use the EGAD evaluation benchmark to evaluate GP-net against two commonly used algorithms, the Volumetric Grasping Network (VGN) and the Grasp Pose Detection package (GPD), on a PAL TIAGo mobile manipulator. In contrast to the state-of-the-art methods in robotic grasping, GP-net can be used for grasping objects from flexible, unknown viewpoints without the need to define the workspace and achieves a grasp success of 54.4% compared to 51.6% for VGN and 44.2% for GPD. We provide a ROS package along with our code and pre-trained models at https://aucoroboticsmu.github.io/GP-net/.