混合联邦-粒子滤波在组合导航系统中的应用  

Mixture Federated-Particle Filter and its Application on Integrated Navigation System

作　　者：李天伟[1] 孙少杰[1] 彭伟骅 郭姣[1] 

机构地区：[1]海军大连舰艇学院航海系 [2]海军驻上海沪东中华造船有限公司军事代表室

出　　处：《计算机仿真》2013年第4期13-16,共4页Computer Simulation

基　　金：国家自然科学基金资助项目(60572160);海军大连舰艇学院科研发展基金资助项目(20122412)

摘　　要：为了解决非线性、非高斯复杂环境下的组合导航系统状态和参数估计问题,针对联邦滤波处理非高斯、非线性系统的不足,提出将粒子滤波引入联邦滤波结构中,形成新的混合联邦-粒子滤波方法。以惯性导航系统/北斗/罗兰C组合导航系统为对象,设计了混合联邦-粒子滤波器,并对所提滤波方法在惯性导航系统/北斗/罗兰C组合导航系统中的应用进行了仿真。仿真结果表明,改进方法能够有效提高系统精度,从而拓宽了联邦滤波的应用范围。In order to solve the integrated navigation system＇s state and parameters estimation problem in the com- plex environment of non - linear, non - Gaussian, a new method combined the advantages of both federated and par- ticle filter has been proposed in this paper to overcome the deficiency of federated filter in dealing with non - Gaussi- an and non - linear system. Taking the INS/BD/Loran - C Integrated Navigation System as a research subject, this paper designed a mixture federated - particle filter and researched the application of the proposed method in the INS/ BD/Loran- C integrated navigation system. The simulation results demonstrate the effectiveness of the proposed method, thus broadening the application range of federated filter.

关 键 词：联邦滤波 粒子滤波 混合系统 组合导航 

分 类 号：TP14[自动化与计算机技术—控制理论与控制工程]