Human Initiated Grasp Space Exploration Algorithm for an Underactuated Robot Gripper Using Variational Autoencoder

TL;DR

This paper introduces a novel grasp space exploration algorithm for underactuated robot grippers that leverages variational autoencoders and expert data to achieve high success rates in simulation.

Contribution

It presents a new mixed analytic and data-driven method for grasp planning using variational autoencoders trained on expert grasps, improving reliability.

Findings

Achieved a 99.91% grasp success rate in simulation.

Effective for multiple objects with limited expert data.

Demonstrates potential for real-time grasp planning.

Abstract

Grasp planning and most specifically the grasp space exploration is still an open issue in robotics. This article presents an efficient procedure for exploring the grasp space of a multifingered adaptive gripper for generating reliable grasps given a known object pose. This procedure relies on a limited dataset of manually specified expert grasps, and use a mixed analytic and data-driven approach based on the use of a grasp quality metric and variational autoencoders. The performances of this method are assessed by generating grasps in simulation for three different objects. On this grasp planning task, this method reaches a grasp success rate of 99.91% on 7000 trials.