Learning to Model the Grasp Space of an Underactuated Robot Gripper Using Variational Autoencoder

TL;DR

This paper introduces a data-driven approach using variational autoencoders to model and explore the grasp space of a multi-fingered adaptive robot gripper, enabling the generation of new grasp configurations.

Contribution

It presents a novel application of variational autoencoders for grasp space modeling in underactuated robotic grippers, based on limited expert grasp data.

Findings

Successfully learned grasp features in a compact latent space

Generated new grasp configurations beyond the training set

Enhanced grasp exploration for complex objects

Abstract

Grasp planning and most specifically the grasp space exploration is still an open issue in robotics. This article presents a data-driven oriented methodology to model the grasp space of a multi-fingered adaptive gripper for known objects. This method relies on a limited dataset of manually specified expert grasps, and uses variational autoencoder to learn grasp intrinsic features in a compact way from a computational point of view. The learnt model can then be used to generate new non-learnt gripper configurations to explore the grasp space.