An accurate and high-sampling-rate GPS trajectory dataset is essential to many trajectory-based applications. However, in practice the GPS dataset always has large position errors and relatively low sampling rate. In this work, we propose and implement a weighting-based map matching algorithm and an interpolation algorithm to calibrate the erroneous and low-sampling-rate GPS trajectory dataset. Moreover, a visualization tool is implemented to help analyze the GPS dataset and evaluate the algorithm’s performance. Map matching algorithms integrate the positioning data and digital road network data to identify the road link on which the vehicle is travelling and the vehicle location on that link. The proposed weighting-based map matching algorithm considers (1) the geometric and topologica...[ Read more ]

An accurate and high-sampling-rate GPS trajectory dataset is essential to many trajectory-based applications. However, in practice the GPS dataset always has large position errors and relatively low sampling rate. In this work, we propose and implement a weighting-based map matching algorithm and an interpolation algorithm to calibrate the erroneous and low-sampling-rate GPS trajectory dataset. Moreover, a visualization tool is implemented to help analyze the GPS dataset and evaluate the algorithm’s performance. Map matching algorithms integrate the positioning data and digital road network data to identify the road link on which the vehicle is travelling and the vehicle location on that link. The proposed weighting-based map matching algorithm considers (1) the geometric and topological information of the road network and (2) historical and future trajectory information. Four criteria are designed to locate a given vehicle position on a map. The interpolation algorithm identifies the path between consecutive GPS points and estimates the vehicle statuses (locations and timestamps) along the path. Both the map matching algorithm and the interpolation algorithm are evaluated in terms of correctness and computational efficiency.