As various driving automation system (DAS) are commonly used in the vehicle, the over-trust in the DAS may put the driver in the risk. In order to prevent the over-trust while driving, the trust state of the driver should be recognized. However, description variables of the trust state are not distinct. This paper assumed that the outward expressions of a driver can represent the trust state of him/her-self. The explicit behaviors when driving with DAS is seen as those outward expressions. In the experiment, a driving simulator with a driver monitoring system was used for simulating a vehicle with the adaptive cruise control (ACC) and observing the motion information of the driver. Results show that if the driver completely trusted in the ACC, then 1) the participants were likely to put their feet far away from the pedals; 2) the operational intervention of the driver will delay in dangerous situations. In the future, a machine learning model will be tried to predict the trust state by using the motion information of the driver.
Explicit behaviors affected by driver's trust in a driving automation system
2019
6 pages, 9 figures, accepted by the 5th International Symposium on Future Active Safety Technology toward Zero Accidents (FAST-zero-19)
Preprint
Electronic Resource
English
Driving behavior model considering driver's over-trust in driving automation system
ArXiv | 2018
|Driver's Lateral Control Strategy as Affected by Task Demands and Driving Experience
SAE Technical Papers | 1977
|DRIVING AUTOMATION & CHANGED DRIVER'S TASK - EFFECT OF DRIVER-INTERFACES ON INTERVENTION
British Library Conference Proceedings | 2016
|Situational Awareness, Driver’s Trust in Automated Driving Systems and Secondary Task Performance
SAE Technical Papers | 2019
|