GA-tuned fuzzy logic control of knee-FES-ergometer for knee swinging exercise

Abstract

Knee-FES-ergometer for knee swinging exercise is introduced as a hybrid exercise for restoration of function of the knee for stroke patients through the application of functional electrical stimulation (FES). The aim of the new knee-FES-ergometer is to provide high intensity knee swinging exercise. It is able to reduce required electrical stimulation and will able to elongate the exercise duration while avoiding early muscle fatigue. Fuzzy logic control (FLC) is used to control the knee trajectory for the purpose of smooth knee swinging exercise. However, conventional FLC rely on human experiences and trial and error for parameter identifications. In this work, a genetic algorithm (GA) is used to tune the FLC to maintain a smooth swinging exercise. The performance of the proposed GA tuned FLC is compared with a manually tuned FLC. Results shows that the GA tuned FLC offers encouragingly better performance.