Research Article
HAU\(\mathbf {M^3}\): A Height Aware Urban Map Matching Mechanism
@INPROCEEDINGS{10.1007/978-3-031-63989-0_27, author={Jie Tang and Sunjian Zheng and Bo Yu and Shaoshan Liu}, title={HAU\textbackslash(\textbackslashmathbf \{M\^{}3\}\textbackslash): A Height Aware Urban Map Matching Mechanism}, proceedings={Mobile and Ubiquitous Systems: Computing, Networking and Services. 20th EAI International Conference, MobiQuitous 2023, Melbourne, VIC, Australia, November 14--17, 2023, Proceedings, Part I}, proceedings_a={MOBIQUITOUS}, year={2024}, month={7}, keywords={Map matching Driving scene classification Images and sensor data Height awareness}, doi={10.1007/978-3-031-63989-0_27} }- Jie Tang
Sunjian Zheng
Bo Yu
Shaoshan Liu
Year: 2024
HAU\(\mathbf {M^3}\): A Height Aware Urban Map Matching Mechanism
MOBIQUITOUS
Springer
DOI: 10.1007/978-3-031-63989-0_27
Abstract
Map matching is an essential component of urban intelligent transportation, providing fundamental data for technologies such as path planning, traffic analysis, and trajectory analysis. However, our commercial deployments showed that existing map matching algorithms behave ineffectively in complex urban scenarios such as elevated and interchange roads due to their insensitivity to the road height change. To improve the map matching for urban map services, we proposed and deployedHAU(\mathbf {M^3}), a novel Height Aware Urban Map Matching Mechanism with Spatio-Temporal Correlation. Firstly, starting from the dynamic driving process of vehicles, an urban driving scene classification model is designed based on the images and sensor data generated synchronously by the vehicles. This model correlates the spatio-temporal state changes of vehicles on roads of different heights. After that, according to the characteristics of different types of map matching algorithms, the classification model is fused into traditional map matching algorithms, enabling them to aware roads at different heights and improving their performance. Evaluation results show that this mechanism significantly improves the accuracy of different types of map matching algorithms in complex urban roads, with an average accuracy increase of over 10%, enhancing the usability of traditional map matching algorithms.
