Underactuated dexterous robotic grasping with reconfigurable passive joints

TL;DR

This paper introduces reconfigurable passive joints for underactuated robotic grippers, enabling complex dexterous manipulation through reconfiguration and learning from single examples, with high success rates on standard datasets.

Contribution

It presents a novel RP-joint design and a grasp learning approach that automatically configures joints for dexterity from minimal examples.

Findings

Achieved 80% success rate on IKEA objects.

Achieved 87% success rate on YCB dataset.

Demonstrated effective reconfiguration for dexterous tasks.

Abstract

We introduce a novel reconfigurable passive joint (RP-joint), which has been implemented and tested on an underactuated three-finger robotic gripper. RP-joint has no actuation, but instead it is lightweight and compact. It can be easily reconfigured by applying external forces and locked to perform complex dexterous manipulation tasks, but only after tension is applied to the connected tendon. Additionally, we present an approach that allows learning dexterous grasps from single examples with underactuated grippers and automatically configures the RP-joints for dexterous manipulation. This is enhanced by integrating kinaesthetic contact optimization, which improves grasp performance even further. The proposed RP-joint gripper and grasp planner have been tested on over 370 grasps executed on 42 IKEA objects and on the YCB object dataset, achieving grasping success rates of 80% and 87%, on IKEA and YCB, respectively.