Realization of robust real time robotic arm control system based on EMG signal

Vast variety of EMG signal applications have been proposed and practiced on EMG control system such as rehabilitation and prosthetic hand and so on. In this paper, a realization of a robust real time robotic arm control system is proposed. First, the root mean square (RMS) of EMG signal for a hand movement is measured and the measured EMG signal is transformed to a complex number. Secondly, the transformed signal is expand Discrete Fourier series (DFS) and the coefficients of DFS is defined the individual feature of hand movement. Thirdly, a hand movement can be recognized by evaluating the Euclidian distance between the coefficients of DFS for unknown hand movement and the reference feature which has been evaluated in the training process. Finally, the experimental results show that the proposed method is robust and faster than the conventional methods.