Electroencephalogram-based Driver Emotional State Detection with Manifold Learning

The detection of the driver emotional state is very important as it is closely related to driving safety. With the development of electroencephalogram (EEG) signal processing and deep learning, researchers are now studying emotional states through EEG signals. However, currently related studies focus less on specific driving tasks, and the model frequently adopted are relatively simple, having limited feature extraction abilities. This paper is dedicated to the study of detecting driver emotional states based on EEG, focusing on a task containing seven categories of driver emotional states, which is relatively complex. We propose a manifold learning model that incorporates feature extraction of symmetric positive definite (SPD) matrix manifold. Experimental results demonstrate that our method outperforms the baselines in detecting various emotional states during driving. This study is useful for detecting human factors in driver emotional state and also helps to prevent traffic accidents caused by negative emotions, thus contributing to the intelligent traffic safety systems.