Proceedings of the Annual Meeting of the Cognitive Science Society

The world is witnessing an increase in the need of technology for human assistance and some of the most prominent may be due to the surge in numbers of elderly needing assistive care, a surge in the numbers of people without mobility either due to accidents, disease or age and the surge in assistive safety technologies like driver assistance. Although asynchronous electroencephanography(EEG) can play a key role in providing assistive technologies in all three of the mentioned areas, the focus of this research is the development of technologies that utilize cognition-in-the-loop to improve living situations of individuals and the community. We have developed a hybrid Brain Computer Interface that implements a signal classification system based on Riemannian Geometry to classify the measurements from the scalp of the user and generate the necessary commands to enable the smart assistive systems to function. The hybrid brain computing interface consists of a steady state visually evoked potential subsystem coupled with a P300 cognitive subsystem, both the systems complement each other. The accuracy seen with the SSVEP system was around 92\% in under 3 seconds which the P300 has accuracies of around 85\% in 3 seconds as well. The SSVEP had a higher signal to noise ratio compared to the P300 which is compatible with previous work in these areas. The assistive technologies tested were powering on smart utilities - light, smart mobility - wheelchairs, smart driver assistance systems - sleep detection, elderly smart assistance - fall detection. However the system is quite capable of working with the following systems; smart utilities - fans, heater and cooling systems, smart driver assistance systems - emotion detection, elderly smart assistance - health situation detection. Given in this manuscript are specific results of the tests on a wheelchair.