模态(人机交互)
固定(群体遗传学)
任务(项目管理)
听觉反馈
控制(管理)
计算机科学
听觉刺激
心理学
认知心理学
人机交互
感知
人工智能
工程类
医学
神经科学
系统工程
环境卫生
人口
作者
Fabienne Roche,A. Somieski,Stefan Brandenburg
出处
期刊:Human Factors
[SAGE Publishing]
日期:2018-12-05
卷期号:61 (5): 839-849
被引量:86
标识
DOI:10.1177/0018720818814963
摘要
Objective: We investigated drivers’ behavior and subjective experience when repeatedly taking over their vehicles’ control depending on the design of the takeover request (TOR) and the modality of the nondriving-related task (NDRT). Background: Previous research has shown that taking over vehicle control after highly automated driving provides several problems for drivers. There is evidence that the TOR design and the NDRT modality may influence takeover behavior and that driver behavior changes with more experience. Method: Forty participants were requested to resume control of their simulated vehicle six times. The TOR design (auditory or visual-auditory) and the NDRT modality (auditory or visual) were varied. Drivers’ takeover behavior, gaze patterns, and subjective workload were recorded and analyzed. Results: Results suggest that drivers change their behavior to the repeated experience of takeover situations. An auditory TOR leads to safer takeover behavior than a visual-auditory TOR. And with an auditory TOR, the takeover behavior improves with experience. Engaging in the visually demanding NDRT leads to fewer gazes on the road than the auditory NDRT. Participants’ fixation duration on the road decreased over the three takeovers with the visually demanding NDRT. Conclusions: The results imply that (a) drivers change their behavior to repeated takeovers, (b) auditory TOR designs might be preferable over visual-auditory TOR designs, and (c) auditory demanding NDRTs allow drivers to focus more on the driving scene. Application: The results of the present study can be used to design TORs and determine allowed NDRTs in highly automated driving.
科研通智能强力驱动
Strongly Powered by AbleSci AI