Robotic Wheelchair Commanded by People with Disabilities Using Low/High-Frequency SSVEP-based BCI

Abstract

Brain-Computer Interfaces (BCIs) are systems developed to improve the daily life of people with disabilities. Nevertheless, this assistive technology is yet far away from the patients’ home and the BCI should adapt to the requirements of the user. This paper presents a practical BCI based on Steady-State Visual Evoked Potentials (SSVEP) currently being used by people with disabilities. The BCI was applied to navigate a wheelchair at two ranges of SSVEP frequency, increasing the possibilities of its application. Two stimulation systems were used: checkerboards flickering at low-frequency range and LED flickering at high-frequency range. Five volunteers with disabilities tested the SSVEP-based BCI. They attempted to command the wheelchair in order to accomplish four different navigation tasks. Then, the volunteers answered a questionnaire about their comfort and performance. Average accuracy detections of 54% and 51% were achieved at low- and high-frequency stimulation, respectively. Volunteers reported lesser visual tiredness when high-frequency LED stimuli were used. The flexible BCI system here developed showed that people with disabilities could operate a robotic wheelchair using visual stimuli in two ranges of frequency. Moreover, preliminary results indicated that visual stimuli flickering at high frequency were more comfortable than low frequency stimuli.