Appropriate uncertainty in manual and automated driving
Table of contents
Project description
The project develops computational models for defining driver-, task-, and situation-specific thresholds of appropriate information sampling frequency from the perspective of traffic safety. The models also enable better understanding of individual variability and learning in drivers’ information sampling from the traffic environment. Models simulate driver- and traffic situation-specific factors’ effects on the information sampling requirements of the driver when the driver is driving manually, assisted by driver assistance systems, or sitting aboard a semi-automated vehicle. The computational models enable development of reliable driver state monitoring algorithms, automated vehicles and driver assistance systems, driver training, as well as safer in-car user interfaces for the drivers. The developed models are applicable for defining dynamic information sampling thresholds and for analyzing operators’ behaviours across dynamic tasks, not only limited to driving.
