Surface electromyography signal (sEMG) can reflect nerves and muscles’ motion to a certain degree and has
great practical value in clinical medical as well as in the medical rehabilitation field, such as nerves and muscles disease
diagnosis, muscles function evaluation and artificial limb control, which achieve certain development and is applied in the
joint motion information identification. Based on the characteristics of hemiplegia patients whose one side limbs motion
function is destroyed, this paper researches on the quantitative identification of upper limbs motion based on sEMG to understand
the patients’ motion intention and consequently supply automatic motion control to the patients. Meanwhile, this
paper focuses on feature extraction and quantitative identification of sEMG as a key technology. The MLPs based on the
Bayesian Regulation was used to identify the degree of the elbow joint of the subject, and reduce the disadvantage of the
normal MLPs which is lack of the enough ability of extension towards this complicated Sub-Gaussian random signals.
The research will be helpful to develop the EMG signals into the actual application of the medical robot.