异步联邦UKF的GNSS/SINS/摄影定位组合导航算法仿真  

Simulation of GNSS/SINS/Photogrammetry Inteqrated Navigation using Asynchronous Federal UKF Algorithm

作　　者：周朋进 吕志伟[1] 丛佃伟[1] 贾铮洋 ZHOU Pengjin;LV Zhiwei;CONG Dianwei;JIA Zhengyang(Institute of Geospatial Information,Information Engineering University,Zhengzhou 450001,China)

机构地区：[1]信息工程大学地理空间信息学院,河南郑州450001

出　　处：《全球定位系统》2018年第6期25-32,共8页Gnss World of China

基　　金：国家自然科学基金(41604032);地理信息工程国家重点实验室开放研究基金项目(SKLGIE2015-M-2-5)

摘　　要：在飞行器进近过程中,为了提高组合系统的导航精度,针对传统联邦滤波器对非线性系统模型易导致滤波发散问题.分析了两种导航方式的优缺点,提出了基于卫星导航/惯性导航/摄影测量(GNSS/SINS/Photogrammetry)的组合导航联邦滤波算法,并推导了系统误差模型.该算法取长补短利用联邦无迹卡尔曼滤波器将GNSS定位和摄影定位、定姿精度高的优势对SINS进行在线误差估计.针对多传感器非等间隔数据采样问题,采用时间与量测更新分离的异步非等间隔联邦滤波算法进行信息融合,并对滤波器结构进行改进以减少算法复杂度.仿真实验证明基于联邦UKF的组合导航系统较传统联邦滤波算法位姿精度有明显的提高,且系统鲁棒性也有一定的增强.In the process of aircraft landing,in order to improve the accuracy of the integrated system and solve the problem of the traditional federal filter that the nonlinear system is easy to filter divergence.This paper analyze the advantages and disadvantages of the two system,propose a federal filter algorithm of GNSS/SINS/Photogrammetry,and derived the system’s error model.This algorithm can assist the SINS with the high precision of GNSS and Photogrammetry,utilizing the federal unscented Kalman filter to estimate errors of SINS.For non-equal interval data problems of multi-sensor,the information fusion of asynchronous non-equal interval federal filter algorithm use time and measurement update separation,and the corresponding improvements of the federal filter structure to reduce the complexity of the algorithm.Computer simulation shows that the position and attitude accuracy based on the UKF-FKF has been significantly improved comparison with traditional federal filter algorithm.

关 键 词：组合导航 信息融合 非线性模型 无迹卡尔曼滤波 非等间隔滤波 

分 类 号：P228.4[天文地球—大地测量学与测量工程]