观测域求积分卡尔曼滤波的机载无源定位算法  

Measure Space Square Root Quadrature Kalman Filter for Airborne Passive Location

作　　者：刘学[1,2] 焦淑红[1] 司锡才[1] 

机构地区：[1]哈尔滨工程大学信息与通信工程学院 [2]中国人民解放军91245部队

出　　处：《西安交通大学学报》2011年第5期137-142,共6页Journal of Xi'an Jiaotong University

基　　金：国家"973计划"资助项目(61393010101-1);国防基础科研基金资助项目(K1503060217)

摘　　要：针对机载无源定位系统中存在滤波稳定性差、收敛速度慢、定位精度差等问题,提出一种观测域平方根求积分卡尔曼滤波算法.新算法兼顾了观测域滤波和平方根求积分卡尔曼滤波的优点,将状态矢量中的各个分量自动解耦,分离了可观测项和不可观测项;通过采用Gaussian-Hermit积分规则提高了非线性变换后随机变量参数的估计精度,有效地降低了状态域与观测域之间转换时存在的高阶误差;使用误差协方差阵的平方根代替协方差阵参与递推滤波,在保证数值稳定性的同时提高了算法的运行效率.计算机仿真表明:新算法提高了滤波稳定性、收敛速度和定位精度.To enhance the robustness,increase the convergence rate and improve the locating accuracy of the filter in the airborne passive location,a novel measure space square root quadrature Kalman filter is proposed.The new filter is characterized by both automatic decoupling capability of measure space filtering and nonlinear filtering capability of quadrature Kalman filter,and decouples the observable component and the unobservable component of the state vector automatically.Gaussian-Hermite quadrature rule is adopted to reduce the higher order truncation error of covariance matrix transformation between measure space and state space,then square root of the error covariance is considered instead of the error covariance in filtering to ensure the numerical stability.Simulation results show the more stable performance,higher convergence rate and better locating accuracy of the proposed filter.

关 键 词：机载无源定位 观测域滤波 求积分卡尔曼滤波器 Gaussian-Hermit积分规则 

分 类 号：TN957.51[电子电信—信号与信息处理]