Steady-state somatosensory evoked potentials for brain-controlled wheelchair

This paper presents three-classes concentration task classification method using spatial-frequency feature of steady-state somatosensory evoked potentials (SSSEPs) for control of brain-controlled wheelchair. The feature extraction methods are based on common spatial pattern (CSP) filtering and fast Fourier-transform (FFT) analysis. A classification method is based on linear discriminant analysis (LDA). Three experimental tasks were performed to concentrate on vibration stimuli of left and right hand finger, and a toe; these tasks were associated with three wheelchair commands: turn left, turn right, and move forward, respectively. The vibration stimuli consisted of vibration motor controlled by micro controller unit (MCU). The experiment by simple paradigm was conducted with three subjects aged between 27 and 28 years old. The overall results show that using the spatial-frequency feature can increase the accuracy in classification of three concentration tasks.