一种改进的基于Sage-Husa 的无迹卡尔曼滤波算法  

An improved unscented Kalman filter algorithm based on Sage-Husa

作　　者：叶泽浩 陈传生 常春贺[1] 张达钊 YE Zehao;CHEN Chuansheng;CHANG Chunhe;ZHANG Dazhao(Air Force EarlyWarning Academy,Wuhan 430019,China)

机构地区：[1]空军预警学院,武汉430019

出　　处：《空天预警研究学报》2023年第5期339-343,共5页JOURNAL OF AIR & SPACE EARLY WARNING RESEARCH

摘　　要：基于Sage-Husa的无迹卡尔曼滤波算法(UKF)能实时估计噪声统计特性来应对系统模型和噪声的不准确性,但是存在不能同时估计状态噪声/状态噪声协方差和量测噪声/量测噪声协方差的问题.为此,提出了一种改进的基于Sage-Husa的UKF算法.首先阐述了基于Sage-Husa的UKF算法;其次改用间接量测更新来简化运算,引入强跟踪原理来约束噪声/噪声协方差的更新频率以及量测噪声协方差的估计结果;最后在系统模型、噪声不准确条件下进行了目标的跟踪仿真.仿真结果表明本文所提算法在上述复杂环境下仍能保持对目标的良好跟踪,验证了该算法的有效性.The unscented Kalman filter algorithm(UKF)based on Sage-Husa can estimate the statistical characteristics of noise in real time to deal with the inaccuracy of the system model and noise,but it has difficulty in estimating state noise/state noise covariance and measurement/noise measurement noise covariance at the same time.In order to solve the aforementioned problem,this paper proposes an improved UKF algorithm based on Sage-Husa.Firstly,the UKF based on Sage-Husa is described.Secondly,on this basis,indirect measurement updating is used to simplify the calculation,and strong tracking principle is introduced to constrain the updating frequency of noise/noise covariance,as well as the estimation result of measurement noise covariance.Finally,target tracking simulation is carried out under the condition that the system model,noise are not accurate.The results show that the proposed algorithm can still track the target well under the above complex environment,with the effectiveness of the proposed algorithm verified.

关 键 词：Sage-Husa UKF算法 状态噪声 量测噪声 强跟踪 

分 类 号：TN957[电子电信—信号与信息处理] TN713[电子电信—信息与通信工程]