Evaluation on Diversity of Drivers' Cognitive Stress Response using EEG and ECG Signals during Real-Traffic Experiment with an Electric Vehicle

Abstract

Inference and prediction of drivers' state is an important research topic in relation to the rapidly increasing market in Adaptive Driver Assistance System (ADAS) and autonomous driving system in recent years. Driving stress has been a particular focus in which state the intelligent system can be critically helpful such as in situations of maneuver handovers. In this research, a rich set of real-traffic driving experiment data was used, which included electroencephalogram (EEG) and electrocardiogram (ECG), and vehicle operational data of 40 drivers in an electric vehicle (EV). This paper aims to show that the individual differences in driving stress, in which the driving environment has been regarded as the most influential determinant inducing stress. The entropy of EEG was calculated to show information regarding stress quantitatively. In addition, we revisit the popular assumption on the relationship between driving stress and driving environment, and identify the differences in individual cognitive responses of driving stress as well as gender specifications.