The traditional map-matching algorithms are often ready for Vehicle Navigation Systems, which always request complicated computation model and needcontinuous and close GPS data whose interval is just about 1 second. Whenraising the accuracy, they lower the efficiency, and are not applicable to process the larger scale GPS data. This paper presents an efficient map-matching algorithm which is very applicable to the real-time traffic information system using large-scale Float Car Data (FCD). This approach designs a kind of data structure of road network and introduces the ideas of partitioning the road network into small grids and dividing the structure of the road network into two levels. It creates the index for the path with the small grid and the direction of line-segment in the road. Examples are provided on a large data set for the Beijing area. The paper demonstrates the efficiency of the algorithm in terms of accuracy and computational speed. For the covering abstract see ITRD E134653.
Samenvatting