Title: A hidden Markov model combined with RFID-based sensors for accurate vehicle route prediction

Authors: Ning Ye; Zhong-qin Wang; Reza Malekian; Ru-chuan Wang; Ting-ting Zhao; Darius Andriukaitis; Algimantas Valinevicius; Dangirutis Navikas; Vytautas Markevicius

Addresses: Institute of Computer Science, Nanjing University of Post and Telecommunications, Nanjing 210003, China ' Institute of Computer Science, Nanjing University of Post and Telecommunications, Nanjing 210003, China ' Department of Electrical, Electronic and Computer Engineering, University of Pretoria, Pretoria 0002, South Africa ' Institute of Computer Science, Nanjing University of Post and Telecommunications, Nanjing 210003, China; Jiangsu High Technology Research Key Laboratory for Wireless Sensor Networks, Nanjing 210003, China; Key Lab of Broadband Wireless Communication and Sensor Network Technology of Ministry of Education, Nanjing University of Post and Telecommunications, Nanjing 210003, China ' Institute of Computer Science, Nanjing University of Post and Telecommunications, Nanjing 210003, China ' Faculty of Electrical and Electronics Engineering, Department of Electronics Engineering, Kaunas University of Technology Studentu St. 50 438, LT 51368 Kaunas, Lithuania ' Faculty of Electrical and Electronics Engineering, Department of Electronics Engineering, Kaunas University of Technology Studentu St. 50 438, LT 51368 Kaunas, Lithuania ' Faculty of Electrical and Electronics Engineering, Department of Electronics Engineering, Kaunas University of Technology Studentu St. 50 438, LT 51368 Kaunas, Lithuania ' Faculty of Electrical and Electronics Engineering, Department of Electronics Engineering, Kaunas University of Technology Studentu St. 50 438, LT 51368 Kaunas, Lithuania

Abstract: The road transport of dangerous goods (RTDG) arouses more and more attentions in recent years. Vehicle location devices only based on GPS technology have played an important role on the current market. However, there are obvious shortcomings by using a simple GPS method in the aspect of positioning accuracy and coverage. In the blind area of GPS, a vehicle's route could not be detached in real time, which will lead to manage and follow the tracks of vehicle difficultly. In this paper, we propose an approach based on hidden Markov model (HMM) to provide static predictions on driver routes. Our approach is based on building the probabilistic model through observation of the driver's habits from a map database involving RFID information. Before we predict a vehicle's route, we firstly compute the shortest path from starting point to destination point. Then through this path we could filter some redundant data. Finally, experiments demonstrate that we acquire high prediction accuracy under different periods of traffic conditions by training the HMM.

Keywords: HMM; hidden Markov model; vehicle route prediction; route filtering; RFID sensors; radio frequency identification; vehicle routing; road transport; dangerous goods; hazardous goods; probabilistic modelling; driver habits; driver behaviour.

DOI: 10.1504/IJAHUC.2016.078473

International Journal of Ad Hoc and Ubiquitous Computing, 2016 Vol.23 No.1/2, pp.124 - 133

Received: 24 Mar 2014
Accepted: 03 Nov 2014
Published online: 22 Aug 2016 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article