Hidden Markov map matching based on trajectory segmentation with heading homogeneity

Map matching is to locate GPS trajectories onto the road networks, which is an important preprocessing step for many applications based on GPS trajectories. Currently, hidden Markov model is one of the most widely used methods for map matching. However, both effectiveness and efficiency of conventional map matching methods based on hidden Markov model will decline in the dense road network, as the number of candidate road segments enormously increases around GPS point. To overcome the deficiency, this paper proposes a segment-based hidden Markov model for map matching. The proposed method first partitions GPS trajectory into several GPS sub-trajectories based on the heading homogeneity and length constraint; next, the candidate road segment sequences are searched out for each GPS sub-trajectory; last, GPS sub-trajectories and road segment sequences are matched in hidden Markov model, and the road segment sequences with the maximum probability is identified. A case study is conducted on a real GPS trajectory dataset, and the experiment result shows that the proposed method improves the effectiveness and efficiency of the conventional HMM map matching method.

Automated summary

Map matching is a crucial preprocessing step for many GPS trajectory-based applications. The conventional map matching methods based on hidden Markov model can suffer from decreased effectiveness and efficiency in dense road networks. This study proposes a segment-based hidden Markov model method to improve the performance of map matching by segmenting GPS trajectories and searching for candidate road segment sequences.