Brain-computer interface for smart vehicle: Detection of braking intention during simulated driving

Neuro-driving simulation framework was proposed in this article for studying neural correlates of braking intention in diversified driving situations. In addition, the possibility was investigated that these neural correlates can be used to detect a participant's braking intention prior to the behavioral response. Electroencephalographic (EEG) and electromyographic (EMG) signals were measured from fifteen participants during they were exposed to several kinds of traffic situations in a neuro-driving simulation framework. After that, the novel characteristic feature was extracted from the measured signals to categorize according to whether the driver intended to brake or not. This proposed novel feature consists of readiness potential (RP), event-related desynchronization (ERD) and event-related potential (ERP) as used in a previous study. The prediction performance of braking intention based on the proposed feature combination exhibited superior prediction performance than simple ERP feature used in a previous study.