GPS map match technique matches a series of geographic coordinates to the road network. However, most existing algorithms consider mainly the geometric and topological relationships of the GPS observations, while the information in the historical floating data remains untouched. In this paper, we develop an enhanced spatial-temporal matching (EST-matching) algorithm effective with low frequency GPS data. The proposed algorithm improves the map-matching technique in the following aspects: (1) The proposed algorithm uses the possible minimum travel time to filter out the unrealistic route and build graph of the candidate paths. (2) Considered the initial matching probability between candidates and the historical data of edge to identify the actual route which the vehicle travel. (3) For vehicles driving at intersections, we introduce direction analysis to increase the algorithm accuracy. The developed (EST-matching) algorithm is tested against the stMM algorithm over a series of real-world data to verify its performance at various data collection frequencies and matching radius. Experiments show that the proposed algorithm outperforms the stMM algorithm in terms of matching accuracy based low sampling frequencies, especially in central urban area.

CICTP