Department of Electronic Engineering, Pukyong National University, Korea.

Speech Title: A Real-Time Hybrid Brain-Computer Interface System for Multidimensional Control using Motor Imagery and Eye Closure
Abstract: Currently, electroencephalography (EEG)-based brain-computer interface (BCI) has mainly focused on identifying human intention and producing certain control commands using following methods individually including motor imagery (MI), steady-state visually evoked potentials (SSVEP), P300, eye closure (EC). Some EEG-based BCI systems have adopted combined methodologies, such as a hybrid of MI/SSVEP or MI/P300. However, these previous solutions have a few practical hurdles, such as making users feel tired due to the use of external stimuli (e.g., flashing lights for SSVEP or P300). This paper presents a combined BCI solution by using MI and EC and then a front-wheel drive (FWD) toy car was used to verify our proposed method. The car was controlled by EEG in order to complete a multidimensional task: going forward, stopping, changing direction. An EC signal was used to switch the car between the three different states, and an MI signal was used to turn the car left or right by 90° by imagining hand grasp motions when the system was in the “changing direction” state. The experimental results show that the participants in the study could control the car to switch between different states and turn left/right with an overall efficiency of 0.8268. The proposed system realizes multidimensional control in a suitable and natural way without the need for external stimuli. Thus, the developed hybrid BCI system has the potential to extend current BCI capabilities in practical utility and self-paced control applications.