Performance of a new enhanced topological decision-rule map-matching algorithm for transportation applications

Map-matching problems arise in numerous transportation-related applications when spatial data is collected using inaccurate GPS technology and integrated with a flawed digital roadway map in a GIS environment. This paper presents a new enhanced post-processing topological decision-rule map-matching algorithm in order to address relevant special cases that occur in the spatial mismatch resolution. The proposed map-matching algorithm includes simple algorithmic improvements: dynamic buffer that varies its size to snap GPS data points to at least one roadway centerline; a comparison between vehicle heading measurements and associated roadway centerline direction; and a new design of the sequence of steps in the algorithm architecture. The original and new versions of the algorithm were tested on different spatial data qualities collected in Canada and United States. Although both versions satisfactorily resolve complex spatial ambiguities, the comparative and statistical analysis indicates that the new algorithm with the simple algorithmic improvements outperformed the original version of the map-matching algorithm.