一种基于模糊推理的导航自适应滤波算法  

An Adaptive Filter Based on Fuzzy Inference System in Navigation Algorithm

作　　者：戴卿[1] 叶城睿 马云峰[1,2] DAI Qing;YE Cheng-rui;MA Yun- feng(Department of Architectural Engineering,Chongqing Water Resources and Electric Engineering College Chongqing 402160, China;College of Civil Engineering, Chongqing Jiaotong University, Chongqing 400074, China)

机构地区：[1]重庆水利电力职业技术学院建筑工程系,重庆402160 [2]重庆交通大学土木工程学院,重庆400074

出　　处：《西安航空学院学报》2018年第3期16-19,共4页Journal of Xi’an Aeronautical Institute

基　　金：渝水职院重点项目(K201708;K201712);重庆教委科学研究项目(KJ1735452)

摘　　要：针对GNSS/SINS组合导航中量测噪声统计特性的时变性,以及由此引发的长航时定位精度下降问题,提出了一种基于模糊推理系统的自适应拓展卡尔曼滤波的导航定位算法。该算法考虑模糊推理系统的优势,对量测噪声协方差进行在线调整,以实现滤波结果的最优估计。经过仿真测试与跑车试验,结果表明该算法在GNSS信号先验统计特性不可知的情况下,对较大偏差有明显的纠偏作用,其定位精度能够保证在5m以内,能更好地进行导航定位解算,具有一定的应用价值。Aiming at time-varying characteristics of measurement noise in GNSS/SINS integrated navigation system and the long-time navigation accuracy reduction problem,a navigation algorithm based on fuzzy inference system for adaptive Extended Kalman filter is proposed in this paper.This algorithm adjusts the covariance of measurement noise online to achieve the optimal estimation of the filter result by using the advantages of fuzzy inference system.After the simulation and filed test,the results show that the algorithm can obviously ensure the positioning accuracy of GNSS signals within 5 m.Thanks to the better navigation positioning solution,the algorithm is of the given application.

关 键 词：组合导航 自适应滤波 模糊推理系统 隶属度函数 

分 类 号：TN967.2[电子电信—信号与信息处理] TN713[电子电信—信息与通信工程]