JOINT FUNCTION CONTROL METHOD FOR ROBOTIC GAIT TRAINING OF STROKE PATIENTS

Abbas Ehsani-Seresht, Majid M. Moghaddam, and Mohammad R. Hadian

Keywords

Rehabilitation robots, assist-as-needed, gait phase detection algorithm, fuzzy state machine

Abstract

This paper introduces the joint function control method (JFCM) of rehabilitation robots that allows the patients to participate in gait rehabilitation exercises actively. The JFCM is based on the studies that show that the function of the hip and the knee joints of humans during different phases of gait is similar to that of spring–damper elements. Therefore, the gait disorders of stroke patients are interpreted as improper spring–damper like function of the joints in some phases of gait. The JFCM adds some virtual springs to the hip and knee joints of the patients to correct their function during gait subtasks. Also, the JFCM uses an adaptive algorithm to adjust the stiffness of virtual springs to provide the needed assistive torques for the patient. The JFCM was implemented through an experimental setup on a male hemiplegic stroke patient. The results reveal that the gait pattern of the patient was enhanced while the patient was able to change the spatial path and the timing of his walking pattern. Also, the joint torques were considerably lower than the impedance control method.

Important Links:



Go Back