Situation Analysis and Adaptive Risk Assessment for Advanced Intersection Safety Systems

Current accident statistics show that intersections are still a hotspot for vehicle crashes. Thus, many vehicle manufacturers and research institutes have identified the need for driver assistance at intersections. Unfortunately, intersections are one of the most complex of all traffic-related scenarios, which complicates the development of comprehensive assistance systems in this area. This thesis presents the development of an Intersection Safety System with special focus on...
More Description

Current accident statistics show that intersections are still a hotspot for vehicle crashes. Thus, many vehicle manufacturers and research institutes have identified the need for driver assistance at intersections. Unfortunately, intersections are one of the most complex of all traffic-related scenarios, which complicates the development of comprehensive assistance systems in this area. This thesis presents the development of an Intersection Safety System with special focus on situation analysis and adaptive risk assessment algorithms which are suitable for online implementation in a vehicle. The demonstrator system developed is able to monitor the intersection environment with several on-board sensors and build an appropriate environmental model including behaviors, intentions and interrelations of all traffic participants in the scene. An adaptive risk assessment judges the possible risk for the current driving task and presents the result to the driver through innovative warning strategies. Functional and user acceptance tests show that the system developed is capable of properly assisting the driver with intersection driving maneuvers and is also well accepted by drivers using the system. At the time of completion of this thesis, the underlying prototype demonstrator vehicle was the first on-board Intersection Safety System demonstrator dealing with these most critical intersection scenarios. Bernd Rossler studied applied computer science in the natural sciences from 1996 to 2001 at the University of Bielefeld, Germany. From 2003 to 2007 he worked at the Group Research Electronics department of Volkswagen AG in Wolfsburg, where the work described in this thesis originated. From 2007 to 2009 he worked as project leader at Ibeo Automobile Sensor GmbH in several national and international research projects in the field of Driver Assistance Systems. Since 2010 he has been working in the research department of SICK AG in Hamburg and acts as coordinator of the EC-funded project INTERSAFE-2 on the topic of Cooperative Intersection Safety Systems.