Home > Published Issues > 2018 > Volume 7, No. 2, March 2018 >

Navigation of an Electric Wheelchair Using Electromyograms, Electrooculograms, and Electroencephalograms

Chiharu Ishii 1, Shunsuke Murooka 2, and Minato Tajima 2
1. Department of Mechanical Engineering, Hosei University, Tokyo, Japan
2. Department of Mechanical Engineering, Hosei University Graduate School, Tokyo, Japan

Abstract— This paper discusses a navigation of an electric wheelchair (EWC) using bio-signals. As bio-signals, electromyogram (EMG), electrooculogram (EOG) and electroencephalogram (EEG) are considered. In navigation of the EWC using EMG, by taking degree and duration of the muscular activity into consideration for navigation of the EWC, travel time of the EWC was reduced. This enhances the maneuverability of the EWC. In navigation of the EWC using EOG, voluntary blink and involuntary blink were distinguished by using support vector machine (SVM). As a result, the average distinction accuracy for voluntary blink of 98.9% was obtained. This implies that the safety to stop the EWC is enhanced. In navigation of the EWC using EEG, operator's EEGs of concentrated state and of relaxed state were distinguished using the SVM. Then, forward moving of the EWC when the operator's EEG of concentrated state was recognized, was successfully achieved. In addition, navigation method of the EWC based on Steady State Visual Evoked Potentials (SSVEP) was proposed.

Index Terms—electric wheelchair, electromyogram (EMG), electrooculogram (EOG), electroencephalogram (EEG), brain-machine-interface (BMI)

Cite: Chiharu Ishii, Shunsuke Murooka, and Minato Tajima, "Navigation of an Electric Wheelchair Using
Electromyograms, Electrooculograms, and Electroencephalograms," International Journal of Mechanical Engineering and Robotics Research, Vol. 7, No. 2, pp. 143-149, March 2018. DOI: 10.18178/ijmerr.7.2.143-149