Study on Real-Time Control of Exoskeleton Knee Using Electromyographic Signal

This paper is concerned with control method for exoskeleton in real-time by using electromyographic signal (EMGs). EMGs is collected from normal subjects when they move their knee flexion-extension in the sagittal plane. The raw EMGs is processed and then input to a four-layer feed-forward neural network model which uses the back-propagation training algorithm. The output signal from neural network is processed by the wavelet transform. Finally, the control orders are passed to the motor controller and drive the exoskeleton knee move by the same way. In this study, the correlation coefficient is used to evaluate the effects of neural network prediction. The experimental results show that the proposed method can accurately control the movement of the knee joint.