An Offline Matching Method for Large Scale Trajectories

Abstract

The proliferation of GPS trajectories collected from vehicles induce the era of LBS. Working as the fundamental preprocess step for many location based services, map matching problem is still an important but challenge research topic. In this paper, we aim to improve the accuracy of offline map matching for large scale trajectories collected from vehicles such as taxis, busses, coaches etc. In our algorithm, spatial-temporal features between the sampling points are analyzed, properties of roadmap are used and driver's route chosen preference is introduced as replenishment to support the temporal-spatial transition between points. The experimental results show that the proposed method can improve the accuracy of the matching, especially in the case of low sampling rate GPS point matching.