Scanning and Locating Target Position with Minimal Training for Brain Computer Interface

Authors

  • M. S. Adha Faculty of Electrical Engineering, Universiti Teknologi Malaysia, 81310 UTM Johor Bahru, Malaysia.
  • N. Mat Safri Faculty of Electrical Engineering, Universiti Teknologi Malaysia, 81310 UTM Johor Bahru, Malaysia.
  • M. A. Othman Faculty of Electrical Engineering, Universiti Teknologi Malaysia, 81310 UTM Johor Bahru, Malaysia.
  • I. Ariffin Faculty of Electrical Engineering, Universiti Teknologi Malaysia, 81310 UTM Johor Bahru, Malaysia.

Keywords:

BCI, EEG, Single Channel, Training,

Abstract

Elicited brain signal recorded noninvasively from the human scalp has been shown sufficiently to control a mobile system to some degree. To further extend the existed human-robot control strategy, this study demonstrated real time target selection based on single channel Brain Computer Interface (BCI) with single trial and without subject training. Subjects were required to synchronously imagine a star rotating and mind relaxation at a specific time and direction. The imagination of a star would trigger a mobile robot suggesting that there was an object to be selected at certain direction in the environment bounded by the human field of vision. This study statistically verifies that there was a significant difference (p<0.001) between star rotation imagery and mental relaxation means. Results from the current study also demonstrate that 95% of subjects were capable of synchronizing the suppression of alpha rhthym at a specific time, which suggested feasibility of the proposed target selection based on scanning scheme. A total of 80% of subjects were capable of selecitng a predetermined target's direction at the first attempt by utilizing individual's optimum parameter. On the other hand, 15% of subjects failed at the first attempt were observed to have succeeded by consecutive trials. Target could not be identified for all trials and six parameters by only a subject. The majority of subjects who could control the proposed system suggest the feasibility of the proposed control strategy by using only F8-electroencephalographic channel and with minimal training.

Published

2017-12-04

How to Cite

Adha, M. S., Mat Safri, N., Othman, M. A., & Ariffin, I. (2017). Scanning and Locating Target Position with Minimal Training for Brain Computer Interface. Journal of Telecommunication, Electronic and Computer Engineering (JTEC), 9(3-9), 21–26. Retrieved from https://jtec.utem.edu.my/jtec/article/view/3137