In the last years many efforts were made to push the trend to highly automated driving from a highway function to urban areas (urban chauffeur functions). With this, new challenges came into focus. One of these challenges is the positioning of the ego car to get lane accurate driving in cities.Due to multipath signal propagation, reflections and even blackouts, a single GPS solution will not be sufficient to face the challenges for a high precise localization and driving. Differential GPS systems (DGPS) combined with inertial measurement units (IMU) are a step forward on this, but currently they are expensive and at longer blackouts the system will fail [8] either. For this reason, further methods were proposed and are under research until today. One big research field is the landmark based localization and derivatives of it. In this paper we introduce a precise localization system that uses series vehicle sensors. We show, that with the use of a set of multiple cameras known from surround view systems, standard GPS equipment and the car Odometry sensors, localization problems inside urban areas could be solved. To estimate the position, lanes Information that have been detected by a set of wide angle cameras mounted around the car are used. These cameras are able to detect and measure the relative position of lane segment markings, arrow markings, pedestrian crossings and stop lines. The information is combined with enhanced data based on open street map (OSM). Thereby a landmark based estimation could be established. Furthermore, the Information and signals from the car sensors and antennas like Odometry and GPS are used for an improved position update and estimation. All Information are finally merged with a particle filter. The main goal of this system is to combine all Information the car could provide by its serial hardware in a cost-effective method for a stable and precise localization in challenging surroundings. The potential performance of the system is evaluated with different driving tests.

Dr. Hadj Hamma Tadjine, IAV GmbH, GERMANY Mr. Amine Kaddache, IAV, GERMANY