An Improvement on the Topological Map Matching Algorithm at Junctions: A Heuristic Approach

Nowadays, there is a growing demand for transportation and location-based services. The rapid progress in wireless and positioning systems caused an ever-increasing use of these systems in vehicles and transportation. Navigation of vehicles relies on matching received positions by Global Positioning System (GPS) or other sensors with the map of road networks to show the user which part of the road they are on. There is a possibility that the algorithm can't recognize the correct link out of the candidate links due to errors in positioning sensors, digital maps, and map matching algorithms. Location-based services, intelligent transportation systems, and users may be misled by incorrect road detection. By combining a topological map-matching algorithm with the Analytic Hierarchy Process (AHP) optimization method, a compound method has been devised. As the material of the study, we have used Garmin GPS data and a 1:2000 urban map of the national cartographic center. We conducted a case study in a dense part of Tehran City in order to test the efficiency of the algorithm. There are three components to the algorithm, one being an initial map match, two being a mapping on a link, and three being a mapping at a junction through the AHP method. The algorithm has been executed in a dense urban network. Because of the presence of high buildings in urban areas we have the most errors in this area. From 906 positioned points the link has been successfully realized in 97.3% of cases. The results are acceptable, and in 2.7% of the remaining cases, error in the positioning system is responsible for the error and it is recommended to improve positioning system errors.