Validation of Grip Force Measurement Using Hand Rehabilitation Robot for Impedance Control

Abstract

Recently, a novel hand rehabilitation robot was developed in an authors’ laboratory for the rehabilitation of patients with neurological disorders such as strokes. In terms of mechanism this robot has features not only to provide full grasping motion by using one motor but also to have finger length adjustment for various people. Although the robot has aforementioned advantages, for impedance control, force information sensed from F/T sensor is not validated yet. Since it is required to figure out the direct force of hand during grasping and opening motion, the reliability of force information needs to be confirmed for applying impedance control to our hand robot. Using simulation tool, Autodesk ForceEffect™, force analysis of the hand robot is implemented. Under three different MCP angles such as 0°, 30°, and 60°, known force is applied at the end-point of the hand robot and torque at F/T sensor can be calculated considering gear ratio. Then, under same condition experiment is carried out. In this case, force is applied using digital dynamometer and torque is directly measured by F/T sensor. As a result, at all cases except for MCP angle 0°, analysis and experiment result have highly linear relation (R^2=0.9916,R^2=0.9917,at MCP angle 30° and 60°). By this relationship, fingertip force can be estimated approximately. Therefore, the foundation of impedance control is established and it leads to the expectation that the impedance control based hand robot will provide the therapy compliant to the patients more precisely, more comprehensively, and much safely. ⓒ 2015 DGIST