We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Advanced Topological Map Matching Algorithm Based on D-S Theory.
- Authors
Zhao, Xiangmo; Cheng, Xin; Zhou, Jingmei; Xu, Zhigang; Dey, Nilanjan; Ashour, Amira S.; Satapathy, Suresh Chandra
- Abstract
Map matching combining vehicle Global Positioning System (GPS) with road attribute information has a significant impact on the traffic route guidance. In the current work, an advanced topological map matching algorithm based on the D-S (Dempster-Shafer) theory is proposed to improve the application for the high-density road network. Data cleaning is carried out to delete the apparent errors of original data and to reduce the data sources. Matching features of the GPS point stated four factors, namely the road speed limit, heading, proximity and spatial correlation. These factors are in line with the vehicle travel characteristics. The D-S theory is applied to dynamically estimate the weight of each factor and further to calculate the fusion result in order to increase the proposed algorithm flexibility. An optimization method based on trajectory shape matching is also proposed to filter the candidate point set to enhance the algorithm accuracy. The results established that the data cleaning can eliminate the invalid data up to more than 5% of the collected data; where using the four factors can guarantee the basic demand of the data matching, while the D-S theory and shape matching can effectively improve the matching accuracy. The proposed system achieved an overall algorithm matching accuracy up to 96.5%.
- Subjects
GLOBAL Positioning System; CARTOGRAPHY; TRAFFIC engineering
- Publication
Arabian Journal for Science & Engineering (Springer Science & Business Media B.V. ), 2018, Vol 43, Issue 8, p3863
- ISSN
2193-567X
- Publication type
Article
- DOI
10.1007/s13369-017-2569-0