Development of BCI Based Wheelchair Using Steady State Visual Evoked Potential

Abstract

This paper shows a Steady State Visual Evoked Potential (SSVEP) based Brain Computer Interface (BCI) framework to control a wheelchair in forward, in reverse, left, right and in stop positions. Four diverse glinting frequencies in low recurrence area were utilized to evoke the SSVEPs and were shown on a Liquid Crystal Display (LCD) screen utilizing LabVIEW. The Electroencephalogram (EEG) signals recorded from the occipital district were initially fragmented into 1 second window and elements were removed by utilizing Fast Fourier Transform (FFT). Three distinct classifiers, two in light of Artificial Neural Network (ANN) and one taking into account Support Vector Machine (SVM) were planned and contrasted with yield better exactness. Ten subjects were taken part in the analysis and the precision was figured by considering the quantity of right location delivered while performing a predefined development succession. One-Against-All (OAA) based multiclass SVM classifier indicated preferred exactness over the ANN classifiers.