MODEL IDENTIFICATION AND CONTROL OF AUTOMATIC FINGER EXTENSOR FOR HAND REHABILITATION

Abstract

The neurological injury in stroke patients affects the hand sensorimotor function such that the muscle weakens for voluntary movements especially in the finger extensor muscles. This condition leads to difficulty in the opening and closing of the hands. For some patients, they suffer from clenched fist or in other words, their fingers remain flexed under normal condition. This paper presents the modeling and control of an automatic finger extensor based on iris mechanism for rehabilitation therapy. The objective of this study is to determine the transfer function of the automatic finger extensor machine and to control the position of the combination of the machine’s poles to achieve the desired opening size so that the patient’s fingers can be extended to the required diameter repetitively in the rehabilitation treatment. The method involves a simple and fast approach of adopting the second order transfer function in the z domain and the implementation of Proportional Integral Derivative (PID) controller to control its motor and eventually its poles movement. The effect of the force from the human hand grasping the poles during the exercise is considered as a disturbance. The System Identification and PID Tuner toolboxes in MATLAB environment are utilized to obtain the coefficients of the transfer function and the PID controller gains. The results show a good fit of 72.94 % has been achieved for the system identification. From the hardware experimental test, it can be seen that the system meets the design requirement under the PID controller with steady state errors of 1.7 % and 4 % for the conditions with and without the human grasping force (disturbance) respectively. This means that the automatic finger extensor machine has successfully extended the normally flexed human fingers to the desired diameter under the controller.