HMM-Based Map Matching and Spatiotemporal Analysis for Matching Errors with Taxi Trajectories

Abstract

Map matching of trajectory data has wide applications in path planning, traffic flow analysis, and intelligent driving. The process of map matching involves matching GPS trajectory points to roads in a roadway network, thereby converting a trajectory sequence into a segment sequence. However, GPS trajectories are frequently incorrectly matched during the map-matching process, leading to matching errors. Considering that few studies have focused on the causes of map-matching errors, as well as the distribution of these errors, the study aims to investigate the spatiotemporal characteristics and the contributing factors that cause map-matching errors. The study employs the Hidden Markov Model (HMM) algorithm to match the trajectories and identifies the four types of map-matching errors by examining the relationship between the matched trajectories and the driving routes. The map-matching errors consist of Off-Road Error (ORE), Wrong-match on Road Error (WRE), Off-Junction Error (OJE), and Wrong-match in Junction Error (WJE). The kernel density method and multinomial logistic model are further exploited to analyze the spatiotemporal patterns of the map-matching errors. The results indicate that the occurrence of map-matching errors substantially varies in time and space, with variation significantly influenced by intersection features and road characteristics. The findings provide a better understanding of the contributing factors associated with map-matching errors and serve to improve the accuracy of map matching for commercial vehicles.