Soft Variable Stiffness Joints for Controllable Grasp Synergies in Underactuated Robotic Hands

Abstract

This paper presents the design and implementation of a soft robotic mechanism for joint stiffening to enable synergy control of an underactuated grasper. Typical underactuated grippers exhibit reduced cost and weight and automatic conforming to object shapes relative to fully actuated grippers but have minimal control over grasp shape and typically a smaller size range of graspable objects. These proposed controllable-stiffness joints (CSJs) act to augment the advantages of underactuated grippers by giving additional pose control and by increasing the size range of graspable objects. The CSJs are implemented here in a parallel gripper to show proof-of-concept, and grasping experiments were performed with and without the controllable-stiffness joints demonstrating their improvements to grasping capability. Over the seven objects tested, the CSJs enabled grasping for two objects that the underactuated gripper was unable to grasp, and for the remaining objects increased the distances from the gripper at which it could successfully grasp the objects by an average of 16.8 mm, representing increases in individual object grasp acquisition space from 16% to 600%.