Research On Action Identification And Track Prediction Of Human Upper Limbs Based On Electromyographic (EMG) Signal

The electromyographic (EMG) signal is the electrical manifestation of neuromuscular activation associated with a contracting muscle. It is an exceedingly complicated signal which is affected by the anatomical and physiological properties of muscles. Surface electromyography (SEMG) has been proved to be a successful method of non-invasive measurement of EMG. EMG was picked up from the intact musculature during volitional motion have been suggested and utilized as an effective method to provide control commands for artificial limbs and functional neuromuscular stimulation. A lot of the researches domestic were focused on the mode recognition of EMG to realize multifunctional control of artificial limbs. However, in the single motion mode, the factors such as velocity of movement, amplitude of movement, and position of the arm remain unclear. The relation between EMG and arm dynamics in the presence of movement has not been fully understood due to the complexity of EMG generation and the complicated relation between EMG and motor control system in the presence of voluntary movements. Effective signal feature extraction and accurate function identification are the crucial problem involved in practical prosthesis control. These problems were discussed theoretically and practically in this paper. The major contents of this thesis are as follows:1. The motion state of human elbow joint was identified based on the EMGs. EMGs were collected from the biceps and triceps muscles of normal subjects when they moved their elbow flexion-extension with time-varying loads. The raw EMG signals were processed and the new defined characteristic was picked up. A four-layer feed-forward neural network model , with the characteristic as its input was developed. The weighted values of the model were optimized with the adjusted back-propagation algorithm. By training, the model can map the transformation from the processed EMG signals to the elbow joint angles. The experimental results showed that the maximal error between the joint angle predicted by the network and the actual joint angle measured by the goniometer was very small.