Cooperation Between Vehicle and Driver: Predicting the Driver’s Takeover Capability in Cooperative Automated Driving Based on Orientation Patterns

Abstract

AbstractThis chapter first describes central development steps of cooperative vehicle control before focusing on the cooperation within the vehicle, between driver and co-system. To enable smooth transitions within this internal cooperation, both agents (driver and co-system) need a mutual understanding of the current capabilities for safely executing the driving task. For this purpose, first the model of confidence horizons is briefly introduced, which represents these mutual capability assessments. In the following, the focus of this chapter is on the assessment of the driver’s ability to take over. First, the state of the art of Driver State Monitoring Systems (DSMS) as well as current challenges are presented. Here it is shown that a prediction based purely on driver observation is not yet possible. Therefore, an alternative approach, the diagnostic takeover request (TOR), is presented, which predicts the takeover capability based on the driver’s initial orientation reaction. In the following, two driving simulator studies are presented in which the diagnostic TOR was used for the first time and thereafter the results are presented and discussed. Finally, a brief outlook is given on how both the diagnostic TOR and the concept of confidence horizons will be further developed.