An EEG-Based Multi-Classification Method of Braking Intentions for Driver-Vehicle Interaction

This paper proposes an electroencephalography (EEG)-based classification method to distinguish emergency and soft braking intentions from normal driving intentions. Time-frequency analysis of EEG signals shows that there exist differences between emergency and soft braking intentions. Power spectral density (PSD) values are used as features. Three Support Vector Machine (SVM)-based binary classifiers are developed to recognize three kinds of driving intentions. Results show that the average recognition accuracy of three classes is over 74%, which shows the feasibility of the proposed method. This study has important values in the exploration of neural signatures of different driving intentions and developing assistant driving systems based on the proposed braking intention detection method.