A practical method for motor imagery based real-time prosthesis control

A potential application of BCI is the way of prosthesis control, while the traditional ways are not appropriate for those who got the high SCI (Spinal Cord Injury). In this paper, considering the practicality of BCI based prosthesis control, especially the avoidance of appended devices, the spontaneous EEG of motor imagery was used as the control signals. The motor imagery is limited in pattern variety, so in our work a careful control strategy was designed for controlling prosthesis with elbow, wrist, and hand joint. Three motor imageries including left/right hand, and feet were used in this system. One was used for switching the current controlling joint and the others were for the forward and reverse of the joints rotation. By this method, the real-time controlling of the three joints prosthesis was achieved. This paper described the auditory paradigm of motor imagery for prosthesis control at first and the relevant coefficient was used for signal analysis and recognition. Then the strategy of controlling 3 joints prosthesis was designed and finally the experiments were carried out to evaluate the ability of prosthesis controlling.