UWB/INS Fusion Positioning Algorithm Based on Generalized Probability Data Association for Indoor Vehicle

Inertial Navigation System (INS) and Ultra-wideband (UWB) fusion localization algorithm can overcome their shortcomings and achieve good localization performance for indoor vehicle. In the fusion system, the non-line of sight (NLOS) error of UWB is still a problem to be overcome. This article presents a loosely coupled method based on generalized probability data association (GPDA) to fuse INS and UWB, and on the basis of traditional GPDA, a GPDA based on modified verification gate (MVG) is proposed. The proposed algorithm correlates the difference between INS and UWB position estimation with GPDA algorithm. Finally, the difference is used for amend the INS localization estimation. The algorithm quality is improved. In the UWB part, GPDA is used again to fuse three different position estimation methods, which alleviates the NLOS error and enhances robustness. Simulation and experimental results show that the quality of the proposed algorithm is better.