Upper limb movement pattern recognition based on sEMG

In the field of rehabilitation medicine, the development of upper limb exoskeleton rehabilitation robot is of great significance for reducing the pressure of medical workers and improving the efficiency of patients’ rehabilitation. However, it is a difficult problem to identify the rehabilitation movement mode of patients with high accuracy. Based on this problem, this paper designs 11 kinds of upper limb movements based on Brunnstrom technology. With the help of surface EMG sensors, the surface EMG signals of the upper limb are collected, from which time domain and frequency domain features are extracted for motion pattern recognition. By constructing LASSO-WOA-SVM recognition model, the feature screening and upper limb motion recognition were completed. The experimental results show that the recognition accuracy of this method reaches 93.5%. Therefore, the upper limb motion pattern recognition method based on LASSO-WOA-SVM is feasible and effective.