Measuring Method and Experiments of Knee Joint Moment of Assisted Exoskeleton Based on Neural Network

The research on the knee joint moment measurement method is of great significance for the moment control and efficiency measurement of the assisted exoskeleton. For most of the existing measuring methods are not suitable when wearing the assisted exoskeleton, knee joint moment measuring methods of assisted exoskeleton system based on Back-propagation (BP) neural network and Long short term memory (LSTM) neural network is proposed, in which surface EMG to joint moment mapping model is created. The estimation of knee moment at different load levels is realized. The experiment of human walking with a natural gait is verified. The root mean square error of the prediction is less than 17%, and the correlation coefficient is above 0.95.