一种多模型自适应联邦滤波器及其在INS/CNS/GPS组合导航系统中的应用

A Multi-model Adaptive Federated Filter and It's Application in INS/CNS/GPS Integrated Navigation System

本文介绍一种基于多模自适应估计的联邦滤波器的原理和特点。设计了INS/CNS/GPS组合导航系统的联邦滤波算法 ,并首次将多模自适应估计方法运用到联邦卡尔曼滤波器中。此外 ,联邦滤波器算法中采用自适应调整信息分配系数的方法。仿真结果表明 ,与采用单一模型的联邦滤波算法相比 ,多模自适应方法与联邦滤波方法结合使用能大大提高导航系统的精度和可靠性。

This paper introduces the theory and characteristics of federated filter based on multi model adaptive estimation. A federated filter of INS/CNS/GPS integrated navigation system is designed and multi model adaptive estimation method is used in federated Kalman filter for the first time. Furthermore, an adaptive adjustment of the coefficients of information sharing is adopted in federated filter. The results of simulation show that as compared with single model algorithm, multi model adaptive estimation combined with the federated filter can improve accuracy and reliability of integrated navigation system greatly.