Map Matching based on Conditional Random Fields and Route Preference Mining for Uncertain Trajectories

TL;DR

This paper introduces a map matching algorithm that combines conditional random fields with route preference mining to enhance GPS trajectory accuracy at low sampling rates.

Contribution

It innovatively integrates route preference with CRF-based features to improve map matching accuracy under low-sampling conditions.

Findings

Improved map matching accuracy at low sampling rates.

Effective use of route preference as a supplementary feature.

Enhanced robustness of map matching in sparse data scenarios.

Abstract

In order to improve offline map matching accuracy of low-sampling-rate GPS, a map matching algorithm based on conditional random fields (CRF) and route preference mining is proposed. In this algorithm, road offset distance and the temporal-spatial relationship between the sampling points are used as features of GPS trajectory in CRF model, which can utilize the advantages of integrating the context information into features flexibly. When the sampling rate is too low, it is difficult to guarantee the effectiveness using temporal-spatial context modeled in CRF, and route preference of a driver is used as replenishment to be superposed on the temporal-spatial transition features. The experimental results show that this method can improve the accuracy of the matching, especially in the case of low sampling rate.