Monitoring driver state offers opportunities to improve in-vehicle safety and to better understand driver behavior. Detecting driver states such as distraction, discomfort, inattention or fatigue is crucial to ensuring safety and to providing the right intervention strategy to improve situational awareness. Comfort and trust are crucial in gaining wider acceptance of automated vehicles. Therefore, the development of a closed-loop human-driver digital twin (HDDT) exhibiting human-like driving behavior, where the driver and the automated system interact with constant feedback, is desirable.

Modelling a HDDT begins with the identification and characterization of driving behaviors via telemetric, psychophysiological, and other data. Questions include: (1) characterizing the interactions between task-directed maneuvers and individual driving styles and, (2) how to reconcile predictive accuracy of a monitoring system with its intrusiveness and driver acceptance.

Modelling of driver state and the development of HDDTs requires expertise in sensor data analysis, cognitive modelling, and human factors. This Dagstuhl Seminar will provide a platform to explore these topics through panels, challenge-led sessions, and open debates. Topics will include: situational awareness and task-switching, exploration of different modalities for interventions, sensor fusion, data augmentation, feature engineering, integration of driver state models into human-machine interfaces, and the ethics of partial data consent. As time allows, design approaches and strategies for providing information and feedback cues to the driver that increase a sense of being supported by, as opposed to being managed by, the monitoring system as a means to increase acceptance and effectiveness will be considered.

Creative Commons BY 4.0

Martin Baumann, Christine Boylan, Roderick McCall, and Bruce Mehler

This seminar qualifies for Dagstuhl's LZI Junior Researchers program. Schloss Dagstuhl wishes to enable the participation of junior scientists with a specialisation fitting for this Dagstuhl Seminar, even if they are not on the radar of the organizers. Applications by outstanding junior scientists are possible until November 14, 2025.