自适应高斯—厄米特滤波器  

Adaptive Gauss-Hermite Filter

作　　者：范炜[1,2] 李勇[3] 林波[1,2] FAN Wei;LI Yong;LIN Bo(Beijing Institute of Control Engineering,Beijing 100190,China;Science and Technology on Space Intelligent Control Laboratory,Beijing 100190,China;Qian Xuesen Laboratory of Space Technology,China Academy of Space Technology,Beijing 100094,China)

机构地区：[1]北京控制工程研究所,北京100190 [2]空间智能控制技术重点实验室,北京100190 [3]中国空间技术研究院钱学森空间技术实验室,北京100094

出　　处：《空间控制技术与应用》2019年第6期47-52,59,共7页Aerospace Control and Application

摘　　要：针对带有未知统计特性噪声的非线性系统,提出了一种新型的自适应滤波器——自适应高斯—厄米特滤波器(AGHF),其过程是通过将Sage-Husa噪声估计器推广到非线性系统,得到更为一般的噪声估计的递推形式,它与高斯—厄米特滤波器(GHF)相融合,得到适用于带有未知统计特性噪声的非线性系统的高精度自适应滤波器.仿真结果表明,当非线性系统存在一类未知统计特性噪声(系统噪声或测量噪声)时,与扩展卡尔曼滤波器(EKF)、GHF和自适应扩展卡尔曼滤波器(AEKF)相比,AGHF滤波器可显著提高对噪声统计特性和系统状态的估计精度.A new approach called Adaptive Gauss-Hermite Filter(AGHF)is proposed to estimate the states and parameters of nonlinear systems with unknown noise statistics.In the new algorithm,Sage-Husa noise statistics estimator is extended to noise statistics estimation for the nonlinear system.Then the new noise statistics estimator is combined with Gauss-Herminte filter(GHF)to deliver Adaptive Gauss-Hermite filter.The simulations show that the new filter is more accurate than the extended Kalman filter(EKF),GHF and adaptive extended Kalman filter(AEKF)for the estimation of the state and the noise statistics when the nonlinear system exists unknown noise statistics in the system equations or the observational equations.

关 键 词：非线性滤波 自适应滤波器 高斯—厄米特滤波器 Sage-Husa噪声估计器 

分 类 号：TP202.4[自动化与计算机技术—检测技术与自动化装置] TP391[自动化与计算机技术—控制科学与工程]