Brain-Computer Interface for Smart Vehicle: Detection of Braking Intention During Simulated Driving

Abstract

It is most essential to stop a vehicle in time for assuring a driver's safety. In this study, a simulated driving environment was implemented to study the neural correlation of braking intention in diverse driving situations. We further investigated to what extent these neural correlates can be used to detect a participant's braking intention prior to the behavioral response. A feature combination method was proposed for the enhancement of detection performance and additional classification of emergency braking triggered by stimuli and voluntary braking. It consists of event-related potential (ERP), readiness potential (RP), and event-related desynchronization (ERD) features. Fifteen participants drove a virtual vehicle and were exposed to the diversified traffic situations in the constructed simulator framework, while technical signals (i.e., gas pedal and brake pedal), electroencephalogram (EEG) and electromyogram (EMG) signals were measured. After that, the neural correlation of the measured signals was analyzed. The proposed framework shows excellent detection performance for various kinds of driver's braking intention. Our study suggests that a driver's braking intention is characterized by specific neural patterns of sensory perception and processing, as well as motor preparation and execution, which can be utilized by smart vehicle technology.