UWB/INS紧组合变分贝叶斯自适应滤波算法  

UWB/INS tightly coupled integration algorithm based on variational Bayesian adaptive Kalman filter

作　　者：徐天河[1,2] 王森 代培培[3] XU Tianhe;WANG Sen;DAI Peipei(Institute of Space Sciences,Shandong University,Weihai,Shandong 264209,China;Shandong Key Laboratory of Optical Astronomy and Solar-Terrestrial Environment,Weihai,Shandong 264209,China;School of Integrated Circuits,Shandong University,Jinan 250100,China)

机构地区：[1]山东大学空间科学研究院,山东威海264209 [2]山东省光学天文与日地空间环境重点实验室,山东威海264209 [3]山东大学集成电路学院,济南250100

出　　处：《导航定位学报》2025年第2期1-8,共8页Journal of Navigation and Positioning

基　　金：国家自然科学基金项目(42388102);国家重点研发计划项目(2020YFB0505800)。

摘　　要：针对超宽带(UWB)与惯性导航系统(INS)紧组合在实际环境中面临非线性误差和噪声干扰的问题,提出一种变分贝叶斯自适应卡尔曼滤波(VBAKF)的UWB/INS紧组合导航算法:通过引入变分贝叶斯方法,自适应调整系统噪声统计特性未知情况下的滤波精度,提升滤波性能,并引入高程约束模型,增强高程方向的定位精度;建立UWB/INS紧组合模型,给出VBAKF滤波算法,对比分析VBAKF与传统自适应卡尔曼滤波(AKF)的状态估计性能差异。实验结果显示,VBAKF方法在东、北、天方向的定位精度相比于传统方法可分别提高16.13%、21.43%和6.25%,表明VBAKF方法能显著提高系统状态估计的准确性和可靠性,有效提高UWB/INS组合导航系统在实测环境下的适应能力。Aiming at the problem that the ultra-wide band(UWB)and inertial navigation system(INS)tightly coupled navigation system is susceptible to nonlinear errors and noise interference in practical environments,the paper proposed a navigation algorithm through the tightly coupled UWB and INS based on the variational Bayesian adaptive Kalman filter(VBAKF):the variational Bayesian method was introduced to optimize the performance of the Kalman filter(KF)and adaptively adjust the filtering accuracy with the unknown statistical characteristics of system noise;in addition,in order to enhance the positioning accuracy of the height direction,a height constraint model was introduced;in this paper,the mathematical model of the tightly coupled UWB and INS was established,the VBAKF algorithm was deduced in detail,and the difference in state estimation performance between VBAKF and traditional AKF algorithm was compared and analyzed.Experimental result showed that the positioning accuracy of the VBAKF method in the east,north and up directions would be increased by 16.13%,21.43%and 6.25%,respectively,compared with the traditional methods,which indicates that the VBAKF method could significantly improve the accuracy and reliability of system state estimation,and effectively improve the adaptability of the UWB/INS tightly coupled navigation system in practical environments.

关 键 词：变分贝叶斯 超宽带(UWB) 惯性导航系统(INS) 紧组合 组合导航 

分 类 号：P228[天文地球—大地测量学与测量工程]