反馈粒子滤波在GPS/INS组合导航系统中的应用  

Usage of Feedback Particle Filter algorithm in GPS / INS integrated navigation system

作　　者：秦岭[1] 刘晨曦[1] 

机构地区：[1]武汉轻工大学电气与电子工程学院,湖北武汉430023

出　　处：《武汉轻工大学学报》2015年第4期56-59,64,共5页Journal of Wuhan Polytechnic University

摘　　要：GPS/INS组合导航系统的性能主要是由滤波器的精度和实时性决定,由于组合导航系统的状态维数较高,导致滤波器的计算量大,因而制约了粒子滤波与高斯粒子滤波算法在实际工程中的应用。针对惯性组合导航系统的特点,基于反馈粒子滤波算法,对位置/速度误差进行了滤波估计,同时利用组合滤波器的输出对INS系统的误差进行修正,最后对相应的算法进行了仿真验证。仿真结果表明,与混合高斯粒子滤波算法相比,反馈粒子滤波算法的滤波精度和实时性都有一定的提高。The performance of GPS/INS integrated navigation system will directly be determinedby the precision and real-time performance of the filters. The dimension of the integrated navigation system is so high that the filter needs the large amount of computation. The result is that the application of particle filter and the Gaussian particle filter algorithms is restricted in practice. According to the characteristics of integrated navigation system, errors of position and velocity are estimated based on feedback particle filter. And the use of the integration method allows the integration navigation system correcting the errors of INS. At last the filter algorithm is verified by simulation. The simulation results show that the precision and real-time performance of feedback particle filter can be improved, compared with those of Gaussian particle hybrid filter.

关 键 词：组合导航系统 反馈粒子滤波 高斯粒子滤波 

分 类 号：U666.1[交通运输工程—船舶及航道工程]