Skip to Main Content
Map-matching is to snap a series of GPS (Global Position System) points onto the road segments in the digital map, so as to estimate the actual trajectory of a moving object. It is an elementary process step for many applications, such as traffic flow analysis, fleet management, and driver assistance. In many existing approaches, the distance between GPS point and roads is a main factor for the position determination. From the year of 2011, a new feature called azimuth angle, which shows the instant driving direction of a vehicle, started to be utilized in our map-matching algorithm. So in this paper, we first studied the feasibility of azimuth angle in accuracy enhancement and then proposed a method to utilize it in candidate filtering. The experiment result shows that the proposed method can improve the efficiency as well as the accuracy of map-matching.