基于H_∞的捷联系统初始对准及参数γ选取方法研究  

Research on SINS Initial Alignment and the Methods for the Parameter γ Selection Based on H_∞ Filtering Technique

作　　者：张涛[1] 徐晓苏[1] 

机构地区：[1]东南大学仪器科学与工程学院,江苏南京210096

出　　处：《四川大学学报（工程科学版）》2009年第4期240-245,共6页Journal of Sichuan University (Engineering Science Edition)

基　　金：国家自然科学基金资助项目(50575042);总装预研项目(51309060402);总装预研项目(51309020503)

摘　　要：针对KALMAN滤波在系统模型不确定或噪声统计特性不准确的情况下滤波精度不高甚至发散的问题,引入了H∞滤波技术,介绍了H∞滤波基本理论和递推公式,与KALMAN滤波在本质上做了比较。建立了12维状态量的初始对准模型,在此基础上做了KALMAN滤波和H∞滤波在白噪声条件和有色噪声条件下的对比仿真,为了说明H∞滤波中参数γ的选取对对准速度和精度的影响,还分别选取3个有代表性的参数γ做了对比仿真,仿真结果表明:在白噪声条件下,两种滤波的精度相当,KALMAN滤波相对H∞滤波要稍好些;在有色噪声条件下,KAL-MAN滤波显得有些发散,而H∞滤波明显对有色噪声的抗干扰性强于KALMAN滤波。同时通过对参数γ的适当选取,对系统的平均性能(精度)和鲁棒性能进行一定的平衡。The H∞ filtering was introduced to solve the problems that KALMAN filtering accuracy is not high or even divergent when system model is uncertain or noise statistics characteristics are not accurate.The basic theory and recursive formulas of H∞ filtering were introduced and compared with KALMAN filtering in essence.The initial alignment model of 12-dimensional states was established and simulated comparing with KALMAN filtering under the conditions of white noise and color noise.Three representative paramet...

关 键 词：捷联惯性系统 初始对准 H∞滤波 KALMAN滤波 

分 类 号：V448.2[航空宇航科学与技术—飞行器设计]