A Comprehensive Driver Behavior Model for the Evaluation of Intelligent Intersections

Many traffic problems occur on urban intersections. Advanced Driver Assistance Systems aim to solve these problems. However, their effects cannot be deduced directly from observing normal driving behavior, because adding a system alters the driving task. A model must be used to represent both normal intersection driving and interaction with ADAS. By using the structure of the driving task, it becomes possible to extend driver models beyond single subtasks. Five modeling tools have been reviewed; none fully fit all reviewing criteria. Nevertheless, we select a promising research approach based on a combination of tools, specifically using a Multi-Agent Systems framework. (precies 100 woorden)