标题：Robust GPS/INS/DVL Navigation and Positioning Method Using Adaptive Federated Strong Tracking Filter Based on Weighted Least Square Principle
作者：Xiong, Hailiang; Mai, Zhenzhen; Tang, Juan; He, Fen
作者机构：[Xiong, Hailiang; Mai, Zhenzhen; Tang, Juan; He, Fen] Shandong Univ, Sch Informat Sci & Engn, Qingdao Campus, Qingdao 266237, Peoples R China.
通讯作者：Xiong, Hailiang;He, F
通讯作者地址：[He, F]Shandong Univ, Sch Informat Sci & Engn, Qingdao Campus, Qingdao 266237, Peoples R China.
关键词：Localization; target tracking; positioning; hybrid navigation; Kalman; filter; weighted least square; federated strong tracking filter
摘要：Multi-sensor integrated positioning technique that combines complementary features of the global positioning system (GPS) and inertial navigation system (INS) for navigation in challenging urban environments has been a hot research area. A variety of algorithms have been proposed over the past two decades for this well-studied field. However, with the increasing demands of seamless positioning, traditional GPS/INS integrated technique faces rigorous challenges, especially in GPS-denied environment, where traditional techniques cannot be applied directly. To improve the precision and robustness of the navigation system, a novel hybrid GPS/INS/Doppler velocity log (DVL) positioning method is proposed, which introduces DVL as the reference information to assist the GPS module to correct the divergence error of INS. A new robust adaptive federated strong tracking Kalman filter (RAFSTKF) algorithm is also presented for data fusion, which has the advantage of robustness with respect to the uncertainty of the system model. Meanwhile, we introduce the least square principle and adaptively adjust information sharing factors to obtain the optimal estimation, which can improve the reliability of the overall system. The theoretical analysis and simulation results demonstrate the effectiveness of the proposed hybrid GPS/INS/DVL positioning method based on RAFSTKF. In addition, the tracking performance of the proposed method outperforms that of traditional federated Kalman filter.