Potential Enhancement for Wrong-way Driver Detection using Precise Attribute Information

Map-matching is widely used in automotive navigation systems to locate vehicle positions on a given digital road map. Various map-matching algorithms have been developed focusing on different application needs. Within the Ghosthunter project, a weighting-function-based map-matching algorithm has been developed for detecting wrong-way driving in order to improve road safety, particularly in Autobahn entrance and exit areas. This paper aims at exploring the potential use of lane-level attributes and height data in improving the success rate of the previously presented algorithm. This algorithm performs well in entrance and exit areas to the Autobahn, with a high success rate of 99.5% in identifying the road on which the vehicle is actually travelling. In the enhanced algorithm presented in this paper the weight coefficients used for computing the total weighting score of candidate roads are adjusted with the aid of one or both kinds of these precise data. The results confirmed that the usage of these precise data can effectively help to detect and correct mismatches at junctions and overpasses.