Hybrid EEG-EOG-based BCI system for Vehicle Control

Brain-Computer Interfaces (BCI) has become a medium of communication and interaction for disabled people. Electroencephalography (EEG) signals are one of the most widely used for such BCI systems. Over the past decade or so, electrooculography (EOG) signals have shown tremendous potential to complement EEG based BCI systems. In this paper, we investigate the possibility of a hybrid BCI system, combining the EEG and EOG signals, for remotely controlling a vehicle, such as a wheelchair, using machine learning technique. Motor Imagery (MI) EEG signals and EOG signals are combined to design this robust and computationally faster system. The proposed system is trained and tested on 13 in-house subject's data and it is able to achieve an average accuracy of 87.3%, where, 3 of the subjects produced more than 90% accuracy.