Citation

Meng H, Zhao X, Chen J, Wang B, Yu Z. Transp. Res. F Traffic Psychol. Behav. 2024; 102: 241-259.

Copyright

(Copyright © 2024, Elsevier Publishing)

DOI

10.1016/j.trf.2024.03.003

PMID

unavailable

Abstract

During autonomous driving, the subjective cognition of scenario risks may lead to cognitive discomfort of occupants. This is a new concept of ride comfort for vehicles, which is being included in the performance evaluation of autonomous vehicles. The observation methods, generation mechanisms, prediction methods, and optimization methods of this cognitive discomfort remain to be explored. In order to quantify the risk cognition of passengers and evaluate cognitive comfort, this paper proposes a physiological representation method using overtaking scenarios as an example. Firstly, physiological data of passengers are collected through road test, and 19 physiological indicators are extracted, including 6 galvanic skin response (GSR) indicators, 5 electrocardiogram indicators, and 8 eye movement indicators. Secondly, the physiological indicator set related to the subjective evaluation results of comfort is constructed through correlation analysis. Finally, a multiple linear regression model is established to represent the level of passenger discomfort, with physiological indicators as inputs and subjective comfort indicators as outputs. The indicators that contribute significantly to the output results in the model are the maximum amplitude of GSR, the rise time of GSR, the maximum pupil diameter, and the average fixation time. The validation results indicate that the overall accuracy of the model is 86.90 %.

Language: en

Keywords

Autonomous Driving; Cognitive Comfort; Multiple Linear Regression; Physiological Indicator; Scenario Risk