Vehicle Localization Via Sensor Fusion Using Evidence Theory

Mourad Zribi, Khalid Touil, and Mohammed Benjelloun

Peer Reviewed

Abstract: In this paper a novel method which provides 2-D position of a vehicle is proposed. This method makes use of global positioning system (GPS) measurement as well as odometric data and a digital road map. Here, we propose to fuse GPS and odometer with digital road map by employing the evidence theory to maintain a correct estimate of the location of a vehicle. The goal of the fusion is to exploit the advantages of each sensor while trying to mitigate their individual limitations. It involves combining the measurement results from different sensors in order to provide knowledge that is not evident from individual sources. The objective in using this combination is to reduce the position error in the presence of slippage, satellite masking, or signal degradation due to multi-path effects. Experimental results show the effectiveness of the use of the evidence theory in the positioning system.
Published in: NAVIGATION: Journal of the Institute of Navigation, Volume 56, Number 1
Pages: 23 - 33
Cite this article: Zribi, Mourad, Touil, Khalid, Benjelloun, Mohammed, "Vehicle Localization Via Sensor Fusion Using Evidence Theory", NAVIGATION: Journal of The Institute of Navigation, Vol. 56, No. 1, Spring 2009, pp. 23-33.
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In