A dataset of 40K naturalistic 6-degree-of-freedom robotic grasp demonstrations

TL;DR

This paper introduces a new dataset of 40,000 naturalistic 6-DOF robotic grasp demonstrations collected using a novel human demonstration method, providing valuable data for advancing deep learning-based grasp planning.

Contribution

It presents a rapid data collection method for naturalistic 6-DOF grasps and a large dataset of successful grasps on diverse objects, addressing limitations of previous datasets.

Findings

Dataset contains 40,000 successful grasps on 109 objects

Demonstration method enables rapid collection of naturalistic grasps

Dataset supports deep learning approaches to grasp planning

Abstract

Modern approaches to grasp planning often involve deep learning. However, there are only a few large datasets of labelled grasping examples on physical robots, and available datasets involve relatively simple planar grasps with two-fingered grippers. Here we present: 1) a new human grasp demonstration method that facilitates rapid collection of naturalistic grasp examples, with full six-degree-of-freedom gripper positioning; and 2) a dataset of roughly forty thousand successful grasps on 109 different rigid objects with the RightHand Robotics three-fingered ReFlex gripper.