Abstract—This paper studied a localization problem for a mobile robot, based on chirp spread spectrum (CSS) indoor localization. This problem in localization is caused by the multi-path effect. The multi-path effect error is difficult to correct because the environment cannot be predicted. To solve the multi-path effect error problem, this paper proposes a pattern matching method. The pattern matching method estimates position by using the fact that data measured for nearby positions is highly similar. To gain high localization accuracy we used the extended Kalman filter (EKF) to fuse the results of localization using pattern matching and dead reckoning.

Index Terms—localization, mobile robot, CSS, indoor, multi-path, pattern matching, EKF

Cite: Seokyong Kim and Jihong Lee, "EKF-based Localization of a Mobile Robot using Pattern Matching," Journal of Automation and Control Engineering ,  ,, Vol.1, No.1, pp. 40-44,  March 2013. doi: 10.12720/joace.1.1.40-44