Olfactory Facilitation of Takeover Performance in Highly Automated Driving

• Published in: Human factors Vol. 63; no. 4; pp. 553 - 564
• Main Authors: Tang, Qiuyang, Guo, Gang, Zhang, Zijian, Zhang, Bingbing, Wu, Yingzhang
• Format: Journal Article
• Language: English
• Published: Los Angeles, CA SAGE Publications 01.06.2021; Human Factors and Ergonomics Society
• Subjects: Accidents, Traffic; Automation; Automobile Driving; Between-subjects design; Design; Hearing; Humans; Interfaces; Odor; Odors; Peppermint; Placebos; Reaction time; Reaction Time - physiology; Safety; Sensory stimulation; Sleep; Touch; Vegetables; Wakefulness; sleep; takeover request; olfactory; highly automated driving; driver behavior
• ISSN: 0018-7208; 1547-8181; 1547-8181
• DOI: 10.1177/0018720819893137

Objective This study aims to quantify the impact of olfactory stimulation and takeover modality on the performance of takeovers in conditionally automated driving. Background Takeover requests are important for the safety of automated vehicles. The reaction time and subsequent performance of drivers in the takeover process are crucial for safety. In this study, peppermint was adopted as an auxiliary modality to the tactile and auditory design of takeover requests. Methods Sixty participants took part in the experiment, which required participants to avoid a stalled vehicle after they were awoken from a state of light sleep by a takeover request. Takeover modality (tactile, auditory, and combined) was the within-subjects factor. In the between-subjects design, half of the participants received a peppermint odor stimulation when the takeover request occurred, and the other half received a placebo (air). Results The presence of peppermint odor did not influence the reaction time, but participants did show signs of being more alert afterwards. For the moment of takeover, use of the auditory modality had a significant positive effect on reaction time compared to the tactile conditions. Conclusion Peppermint odor had a positive impact on drivers’ takeover quality when engaged in nondriving-related activities such as light sleep, and the takeover request modalities were shown to be crucial for a safe and successful takeover. Application The results will be useful as a reference for developers of automated driving systems to design human–machine interfaces, shorten the driver’s reaction time, and improve takeover quality.