改进抗差卡尔曼滤波的UWB定位模型  

An improved robust Kalman filtering model for UWB positioning

作　　者：赵佳星 王坚 刘严涛 王志涛 ZHAO Jiaxing;WANG Jian;LIU Yantao;WANG Zhitao(School of Architecture and Urban Planning,Beijing University of Civil Engineering and Architecture,Beijing 100044,China;School of Geomatics and Urban Spatial Informatics,Beijing University of Civil Engineering and Architecture,Beijing 102616,China)

机构地区：[1]北京建筑大学建筑与城市规划学院,北京100044 [2]北京建筑大学测绘与城市空间信息学院,北京102616

出　　处：《导航定位学报》2024年第4期174-180,共7页Journal of Navigation and Positioning

基　　金：北京市自然科学基金面上项目(8222011)。

摘　　要：针对超宽带(UWB)技术已在室内定位、智能交通等领域广泛应用,但由于障碍物遮挡或其他原因,使得UWB测距值可能存在误差,从而大幅度降低UWB的定位精度和可靠性的问题,分析抗差卡尔曼滤波(RKF)在UWB定位中的特点,在此基础上,提出改进的RKF算法(RKF-SW)进行粗差检测,根据预测残差进行第一次样本异常值检验探测,采用1/4样本分位数进行第二次样本异常值探测。实验结果表明,与RKF算法相比,RKF-SW算法具有较强的异常值识别能力,能够提升UWB系统的定位精度,开阔场景下定位精度可提高约80.0%,遮挡物干扰情况下可提高约82.1%。Ultra-wide band(UWB)technology has been widely used in indoor positioning,intelligent transportation and other fields.However,due to the obstructions or other factors,UWB ranging can have errors,which greatly reduces the positioning accuracy and reliability.This paper analyzed the characteristics of the robust Kalman filter(RKF),and improved a RKF-SW(sliding window,SW)for gross error detection.Initial outlier detection was performed based on the prediction residual,and an 1/4 sample quantile was used for subsequent sample outlier detection.Compared with the RKF,the RKF-SW had strong outlier identification capabilities and improved the positioning accuracy.In open scenes,the positioning accuracy was increased by approximately 80.0%,and in the case of interference from obstructions,the positioning accuracy was increased by approximately 82.1%.

关 键 词：异常值 样本分位数 抗差估计 超宽带(UWB) 

分 类 号：P228[天文地球—大地测量学与测量工程]