Poster: A Magnetic Fingerprint-based Automobile Position Tracking using Smartphones

Abstract

The state-of-the-art navigation frequently fails to provide precise outdoor localization especially in urban areas due to GPS errors. Conventional approaches tracking vehicle positions have used the particle filter algorithm, and we focus on magnetic fingerprint features for increasing performance of the algorithm as landmarks. We investigated the magnetic features in two conditions such as in-vehicle and at an intersection. More diverse traffic environments and an effective design of the particle filter algorithm will be complemented in the future research. We expect these magnetic fingerprints significantly improve outdoor localization algorithms especially in urban area.