组合导航的集中式容错卡尔曼滤波算法  

Centralized fault-tolerant Kalman filter for integrated navigation

作　　者：赵琳[1] 康瑛瑶 程建华 吴谋炎 ZHAO Lin;KANG Yingyao;CHENG Jianhua;WU Mouyan(College of Intelligent Systems Science and Engineering, Harbin Engineering University, Harbin 150001, China)

机构地区：[1]哈尔滨工程大学智能科学与工程学院,黑龙江哈尔滨150001

出　　处：《哈尔滨工程大学学报》2021年第6期845-850,共6页Journal of Harbin Engineering University

基　　金：国家自然科学基金项目(61633008);黑龙江省杰出青年基金项目(JJ2018JQ0059);中央高校基本科研业务费专项基金项目(HEUCFP201768).

摘　　要：针对多源信息组合导航系统中集中卡尔曼滤波器存在的斜坡噪声抑制难题,本文提出一种容错卡尔曼滤波算法。该算法在传统极大似然抗差估计的基础上,进一步考虑量测信息的先验概率分布,设计量测信息在线质量评价函数,当量测信息含有阶跃噪声、冲击噪声、及斜坡噪声时实现对量测信息的质量评价,并依据质量评价信息构造量测噪声权函数与自适应因子,通过调节量测信息权重,抑制包括斜坡噪声在内的量测粗差对滤波精度及可靠性的影响,实现容错滤波。将该算法应用于多传感器组合导航系统中进行仿真实验验证,仿真结果证明:本文提出的容错卡尔曼滤波算法可以有效抑制量测信息中的阶跃噪声、冲击噪声、及斜坡噪声,滤波器的可靠性和精度得到了有效提升。The traditional centralized Kalman filter employed by integrated navigation systems with multi-source information is unable to limit the ramp measurement error.In this paper,a fault-tolerant Kalman filter(FTKF),which takes the prior probability distribution into account,is proposed based on the maximum likelihood estimation to restrain step errors,impulse errors,and ramp errors.The quality evaluation method and function are designed to evaluate the measurement availability and the equivalent weight function and adaptive factor are constructed based on the quality evaluation information.By adjusting the weight of measurement information,the influence of gross measurement error including slope noise on the filtering accuracy and reliability is limited and fault-tolerant filtering is realized.The algorithm is applied to the multi-sensor integrated navigation system for simulation and verification.The simulation results show that,when employed by an integrated navigation system,the proposed FTKF can limit the measurement outliers,including step errors,impulse errors,and ramp errors,and improve the filter accuracy and reliability.

关 键 词：集中滤波 容错滤波 自适应滤波 斜坡噪声 质量评价函数 等价权函数 自适应因子 组合导航 

分 类 号：U666.1[交通运输工程—船舶及航道工程]