Journal of Northeastern University Natural Science ›› 2017, Vol. 38 ›› Issue (4): 472-475.DOI: 10.12068/j.issn.1005-3026.2017.04.004

• Information & Control • Previous Articles     Next Articles

Map Matching Algorithm Based On Hidden Markov Model and Genetic Algorithm

WU Gang, QIU Yu-jing, WANG Guo-ren   

  1. School of Information Science & Engineering, Northeastern University, Shenyang 110819, China.
  • Received:2015-12-07 Revised:2015-12-07 Online:2017-04-15 Published:2017-04-11
  • Contact: QIU Yu-jing
  • About author:-
  • Supported by:
    -

Abstract: A new map matching algorithm was proposed on the basis of the hidden Markov model and the genetic algorithm. Firstly, the HMM probability matrix was initialized. Then, the parameters were learned by using the forward-backward algorithm, and a set of road sections was predicted by using the Viterbi algorithm. Finally, taking section sequence as population, the optimal section sequence was obtained by using the genetic algorithm. By using the taxi GPS data from Beijing in 2012 to test the traditional algorithm based on hidden Markov model and the proposed algorithm, the results showed that the traditional algorithm based on hidden Markov model has a matching accuracy below 90% and the proposed algorithm has a matching accuracy above 90%.

Key words: map matching, hidden Markov model, genetic algorithm, matching accuracy, road network data

CLC Number: