Development of a Driver Behavior Framework for Manual and Automated Control Considering Driver Cognition

Abstract

As a crucial component of traffic safety, operational quality, and network performance, driver behavior has been the subject of numerous studies. However, research has focused primarily on descriptive mathematical models of the primary driving tasks (car-following, lane changing), while rarely considering the underlying human factors affecting driver behavior. This quality of existing models means that they are not generally capable of adapting to systemic changes in driving behavior. At the same time, vehicle automation, one of the most revolutionary innovations in the history of transportation, advances at a very rapid pace. This development will result in deep systemic changes in the driver role and behavior, during the unavoidable transition period towards fully automated transportation networks, which the existing descriptive models are ill equipped to predict. To achieve that, additional information about driver behavior derived from the field of cognitive sciences, and psychological constructs like cognitive workload and situational awareness, need to be integrated into driving behavior models in order to describe the driver state under various levels of automation. This research aims to fill that gap by proposing a robust driver behavior framework that takes into account human factors and can be applied to both traditional manual driving, as well as driving of vehicles with varied automation capabilities. Based on a comprehensive literature review, the study proposed an experimental methodology, and a data collection and analysis plan that can validate the behavioral framework for use in future transportation applications.