EMG-based Continuous Control Method for Electric Wheelchair

Abstract

This paper presents a continuous control method of electric wheelchair based upon surface electromyographic signals (EMG), ultimately, for quadriplegics. The proposed method utilizes two EMG signals as inputs for the muscle-computer interfaces (MCI). Since Zygomaticus major muscles located in the right and left sides of human face are able to excise individually and to control contractile forces voluntarily, the surface EMG signals of both muscles satisfy core requirements for the development of EMG-based electric wheelchair control system, such as independent and continuous speed control of two wheels. For this, the envelopes of the signal waveforms are first extracted to reflect the moving average activities by using RMS (root mean squares) operations. Also, in order to obtain the desired linear and angular velocities of the electric wheelchair, the RMS signals are processed sequentially as follows; normalizing the RMS signals and then determining the control inputs of the electric wheelchair. Finally, the effectiveness of the proposed control scheme is verified through several experiments.