Modeling and Control of Functional Electrical Stimulation Cycling Training System

As a rehabilitation technique, functional electrical stimulation cycling training system can improve patients' cardiopulmonary function, enhance muscle strength, and restore some motor function. However, for functional electrical stimulators, the pattern of electrical stimulation in cycling training is usually obtained by physiotherapist based on trial and error or clinical experience, Inability to maintain a training speed for an extended period of time. Therefore, this study proposed a kind of pedal-Hill modeling to establish an optimal stimulus mode with the maximum torque efficiency optimization objective. A particle swarm optimization (PSO) algorithm, back propagation (BP) neural network algorithm and a composite algorithm of proportional integral differentiation (PID) control were used to construct an adaptive controller. We conducted experiments on six subjects to validate the system. First, the six subjects' own stimulation patterns were found separately according to the proposed modeling approach, and then each subject performed three sets of experiments for cycling without electrical stimulation, cycling with fixed pulse width, and cycling with adaptive adjustment of pulse width by the fabricated controller. In the traditional training mode, the cycling speed of the six subjects fluctuated greatly, in the FES-engaged mode, six subjects experienced longer sustained riding and more stable cycling training speeds. RMSE: fixed pulse width < adaptive adjustment of pulse width, AVE: traditional training mode > fixed pulse width >28 adaptive adjustment of pulse width, training time: traditional training mode < fixed pulse width < adaptive adjustment29 of pulse width, number of stops: traditional training mode > fixed pulse width > adaptive adjustment of pulse width. The results of the study showed that the adaptive pulse width control system of the fabricated controller allows subjects to train at a more stable speed compared to the no control and fixed control conditions.