Detection of multi-class emergency situations during simulated driving from ERP

Abstract

We present a driving simulator study investigating whether a driver's braking intention in emergency situations can be detected under more general circumstances than previously described in the literature. Precisely, we here simulated three kinds of realistic emergency situations instead of only one as considered in Haufe et aI., 2011. For each of the three situations, the analysis of electroencephalography (EEG) data reveals a different characteristic spatio-temporal event-related poten tial (ERP) sequence. For all stimuli, topographical maps of area under the curve (AUC) scores related to the discrimination between emergency and normal driving situations show a significant positive deflection in parietal regions about 300ms post-stimulus. Thus, it is possible to predict different emergency situations from EEG before the actual braking. A classification analysis indeed reveals that EEG-based emergency braking detection can be performance faster than electromyography-or pedal-based detection, while being as robust.

Decision of braking intensity during simulated driving based on analysis of neural correlates

“…The over-fitting of training data is thought to contribute to large differences of the prediction results based on partially trained regression model before technical response and that of the prediction results based on combination feature after technical response. On the other hands, previous studies [9,10] show the outstanding performance for detection of resting state. Therefore, it is considered that low decoding performance of partially trained regression model for resting state could be enhanced by combination of these methods (i.e.…”

“…ERP features were extracted in the same way as previous studies [9,10]. On the other hands, additional segmentation for the time interval between -300 ms and 600 ms relative to the stimulus onset was performed to obtain RP features.…”

“…However, these studies were not directly related to driver's safety assurance. Only a few of studies focused on driver's safety by detecting emergency situations during simulated driving [9,10]. The neural correlates related to driver's behavioral responses in emergency situations were investigated in these studies.…”

“…For example, event-related potentials (ERP) were analyzed mainly to detect driver's specific mental state (i.e. neurophysiological responses caused by emergency situations) in the previous studies [9,10]. In this study, three kinds of signal components were combined to robustly decode the driver's braking intention.…”

Recognition of Drivers’ Hard and Soft Braking Intentions Based on Hybrid Brain-Computer Interfaces

“…Biological information includes electroencephalography (EEG) signals and electromyography (EMG) signals [16,17]. Because of the early occurrence of EEG signals, EEG signalbased brain-computer interfaces (BCIs) have been applied to driver behavior research [18][19][20]. Although these BCIs have made great progress in braking intention detection, the detection performance is not stable because of the properties of EEG signals.…”