Abstract
This contribution presents an advanced method to predict the road course at long distances ahead of the vehicle. For Advanced Driver Assistance Systems (ADAS) like a pedestrian warning system, it is essential to know the road course even at large distances ahead of the vehicle. This method uses a digital map as a powerful additional sensor that gives information about the shape of the road and an imaging radar sensor for the localization task. A local map of the vehicle environment accumulates the measurements of the radar sensor. A Monte Carlo based method uses this local map to determine the vehicle pose (position and orientation) on the digital map. The algorithm enables robust road course prediction at long distances.
Keywords: digital map, lane detection, ADASIS