Citation

Lee J, Yang JH. Int. J. Automot. Technol. 2020; 21(3): 719-728.

Copyright

(Copyright © 2020, Holtzbrinck Springer Nature Publishing Group)

DOI

10.1007/s12239-020-0070-3

PMID

unavailable

Abstract

As partially automated driving vehicles are set to be mass produced, there is an increased necessity to research situations where such partially automated vehicles become unable to drive. Automated vehicles at SAE Level 3 cannot avoid a take-over between the human driver and vehicle system. Therefore, how the system alerts a human driver is essential in situations where the vehicle autonomous driving system is taken over. The present study delivered a take-over transition alert to human drivers using diverse combinations of visual, auditory, and haptic modalities and analyzed the drivers' brainwave data. To investigate the differences in indexes according to the take-over transition alert type, the independent variable of this study, the nonparametric test of Kruskal-Wallis was performed along with Mann-Whitney as a follow-up test. Moreover, the pre/post-warning difference in each index was investigated, and the results were reflected in ranking effective warning combinations and their resulting scores. The visual-auditory-haptic warning scored the highest in terms of various EEG indexes, to be the most effective type of take-over transition alert. Unlike most preceding studies analyzing post-take-over-alert human drivers' response times or vehicle behavior, this study investigates drivers' brainwave after the take-over warning.

Language: en