Robust Cooperative Indoor Localization Based on Reliability Evaluation

Abstract

Due to the complex mixed non-line-of-sight (NLOS) and line-of-sight (LOS) propagation conditions, it is a challenging problem to achieve robust and high-performance localization in complex indoor scenarios. To solve this problem, this paper proposes a robust cooperative localization method (CLM) based on the reliability evaluation. The proposed CLM scheme leverages the advantages of the compressive sensing (CS) algorithm in the LOS condition and the particle filter (PF) algorithm in the NLOS condition. We propose a reliability evaluation method to evaluate the estimation results of the CS and PF algorithms, and develop a cooperative localization framework by considering the reliability when merging the estimation results. We also develop a localization system based on ultra-wideband hardware to evaluate the performance of the proposed CLM. Extensive experimental results carried out in different scenarios demonstrate the robust performance of the proposed schemes and the stability of the developed system.