A pedestrian-oriented map matching algorithm for map information sharing systems in disaster areas. (1st January 2012)

• Record Type: Journal Article
• Authors: Asakura, Koichi; Takeuchi, Masayoshi; Watanabe, Toyohide
• Is Part Of: International journal of knowledge and web intelligence. Volume 3:Number 4(2012)
• Page Start: 328
• Page End: 342
• Publication Date: 2012-01-01
• Subjects: map matching -- navigation system -- ad-hoc network -- disaster area; Information retrieval -- Periodicals; Web site development -- Periodicals; Internet programming -- Periodicals; 004.67805
• Languages: English
• ISSNs: 1755-8255

Abstract

In this paper, we propose a pedestrian-oriented map matching algorithm for tracking moving trajectories of pedestrians. Our algorithm is designed for a map information sharing system among refugees in disaster areas. In this situation, refugees have mobile terminals with GPS devices and move to shelters at walking speed. Thus, our algorithm has to be suitable for battery-driven mobile terminals. In order to reduce battery consumption, our algorithm is based on a geometric curve-to-curve matching approach in which computation resources are less required in comparison with other complicated approaches such as probabilistic map matching, statistical map matching and so on. Furthermore, in order to deal with matching errors, our algorithm has following features: initial matching point selection, candidate road segments selection and incremental matching method. We conduct experiments with real time sequence data captured by GPS. Experimental results shows that our proposed algorithm can achieve the best result in comparison with the other conventional geometric map matching methods.