基于OEAR算法优化的卡尔曼滤波定位算法  

Kalman Filter Localization Algorithm Based on OEAR Optimization

作　　者：王丽娟 马刚 张东 吕途 WANG Li-juan;MA Gang;ZHANG Dong;LV Tu(School of Electric Acadame,North China University of Water Resources and Electric Power,Zhengzhou Henan 450000,China)

机构地区：[1]华北水利水电大学电力学院,河南郑州450000

出　　处：《计算机仿真》2020年第9期235-238,332,共5页Computer Simulation

基　　金：河南省科技攻关项目(162102210075);华北水利水电大学高层次人才计划项目(40434)。

摘　　要：卡尔曼滤波算法运用于动态定位过程分为先验估计和后验估计两个过程,后验估计过程中观测值精度限制了卡尔曼滤波定位算法定位精度。传统均值滤波由于其算法复杂度较小,成本较低被广泛使用,但设备因素、人为因素、环境因素等对算法结果影响较大。针对以上问题提出测距误差离群去约束算法优化的卡尔曼滤波定位算法,通过减小观测值输入值误差来提高最终算法定位精度。利用Matlab软件对20条随机路径进行仿真,仿真结果显示,,改进后的算法有0.5m-1m的精度提升。The Kalman filter algorithm is applied to the dynamic positioning process,which is divided into two parts:a priori estimation and a posteriori estimation.The accuracy of the observed value in the posterior process limits the positioning accuracy of the Kalman filter localization algorithm.The traditional mean filtering is widely used be⁃cause of its small algorithm complexity and low cost.However,equipment factors,human factors and environmental factors have a great impact on the algorithm results.Aiming at the above problems,a Kalman filter localization algo⁃rithm based on the outlier error anchor removal algorithm is proposed.The algorithm can improve the positioning accu⁃racy of the final algorithm by improving the input precision of the observed value.The Matlab was used to simulate 20 random paths.The simulation results show that the improved algorithm has an accuracy of 0.5m-1m.

关 键 词：卡尔曼滤波定位算法 均值滤波 误差离群去约束 观测值 

分 类 号：TN929[电子电信—通信与信息系统]