A novel method for elbow joint continuous prediction using EMG and musculoskeletal model

As a representation of muscle activation dynamics, electromyograms (EMG) signals can reflect muscle contraction status. The status has some relationship with body movements under certain circumstance. This paper is aimed at upper limb elbow joint continuous prediction using EMG signals. Unlike the conventional pattern recognition method, a more quantitative relationship between EMG signals and joint angles has been developed using the Hill-based musculoskeletal model. The EMG signals are recorded from biceps muscle and its antagonist muscle, triceps brachii muscle. The movements of upper limb are voluntary elbow flexion and extension in vertical plane and horizontal plane. The computational time consuming of the proposed method is little and it can be implemented in real-time easily. Five subjects participated in the experiment to evaluate the efficiency of this method.