基于UT变换改进多模型滤波的水下航行器导航方法  

Improved multiple model kalman filtering algorithm based on UT to navigation system of underwater vehicle

作　　者：王泽元[1] 许昭霞[2] 

机构地区：[1]中国人民解放军61818部队 [2]中国人民解放军61081部队

出　　处：《舰船科学技术》2014年第10期73-77,共5页Ship Science and Technology

摘　　要：为了提高水下航行器组合导航系统的精度,针对滤波算法存在较大的截断误差和累积误差等问题,提出了一种基于无迹变换改进多模型滤波算法,并利用辅助信息对估计结果进行修正。首先通过无迹变换产生Sigma点对非线性测量方程进行近似,构造伪观测量进行偏差估计,然后利用基于加权因子的辅助信息融合算法,消除累积误差,进一步提高系统估计精度,最后给出算法的实现过程。仿真结果表明:与常规的多模型滤波算法相比,本文方法提高了估计精度。To improve the navigation precision of the underwater vehicle,an improved multiple model kalman filtering algorithm based on unscented transform（ UT） was proposed for the truncation errors and accumulative errors,which used the auxiliary information to refine the estimation results. Firstly,the sigma points coming from UT were adopted to approximate the measurement nonlinear equation and build the pseudo-measurements and estimate the slopes. Then, a data fusion algorithm with sensor weighted coefficients was utilized to amend the accumulative errors. Finally,a process of simplified algorithm is presented. Simulation results show that this algorithm can effectively improve the estimation accuracy compared with the traditional multiple model algorithm.

关 键 词：无迹卡尔曼滤波 状态估计 多模型算法 组合导航 水下航行器 

分 类 号：U666.1[交通运输工程—船舶及航道工程]