摘要
针对室内复杂环境下单一定位技术难以满足高精度定位需求的问题,提出了一种用于非线性系统状态估计的中心差分扩展集员滤波(CDESMF)算法来融合超宽带(UWB)和惯性测量单元(IMU)的测量信息。为了克服泰勒展开的固有缺陷,采用低阶多维斯特林插值公式代替泰勒展开将非线性系统线性化,以提高算法精度并且有效抑制数据的抖动。实验结果表明:本文研究的融合算法定位精度在5 cm以内,比扩展卡尔曼滤波(EKF)融合算法定位精度提高了21%。
Aiming at the problem that it is difficult for a single positioning technology to meet the high-precision positioning requirements in the complex indoor environment,a central differential extended set membership filtering(CDESMF)algorithm for nonlinear system state estimation is proposed to integrate the measurement information of ultra-wideband(UWB)and inertial measurement unit(IMU).In order to overcome the inherent defects of Taylor expansion,the nonlinear system is linearized by using low order multidimensional Stirling interpolation formula instead of Taylor expansion to improve the precision of the algorithm and effectively suppress the data jitter.Experimental result shows that the positioning precision of the studied fusion algorithm is within 5 cm,which is 21%higher than that of extended Kalman filtering(EKF)fusion algorithm.
作者
吴晓烨
何青
李灵
张杏
WU Xiaoye;HE Qing;LI Ling;ZHANG Xing(School of Electrical and Information Engineering,Changsha University of Science&Technology,Changsha 410114,China)
出处
《传感器与微系统》
北大核心
2025年第5期112-114,119,共4页
Transducer and Microsystem Technologies
基金
国家自然科学基金资助项目(62103063)
湖南省教育厅科学研究项目(22B0329)。
关键词
集员滤波
超宽带
惯性测量单元
室内定位
set-membership filtering
ultra-wideband
inertial measurement unit
indoor localization
