基于矩阵分解的捷联惯导多位置系统级标定方法  

Multi-Position Systematic Calibration Method for SINS Based on Matrix Decomposition

作　　者：赵桂玲[1] 谭茂林 吴偲 ZHAO Guiling;TAN Maolin;WU Cai(School of Geomatics,Liaoning Technical University,Fuxin Liaoning 123000,China)

机构地区：[1]辽宁工程技术大学测绘与地理科学学院,辽宁阜新123000

出　　处：《传感技术学报》2022年第6期725-731,共7页Chinese Journal of Sensors and Actuators

基　　金：辽宁省自然科学基金项目(2020-MS-303,2020-BS-259);辽宁省教育厅一般项目(LJ2020JCL015);国家级大学生创新创业训练计划项目(202110147010)。

摘　　要：针对捷联惯导系统级标定中滤波法路径设计复杂、误差参数辨识未进行深入的理论分析等问题,提出一种基于矩阵分解的系统级标定算法。将系统级标定中的拟合法路径设计引入滤波法,对静止和转动过程中的误差参数辨识进行深入的理论推导和分析,设计一组7位置连续标定路径。基于矩阵分解将安装误差角分解为非正交角和正交的失准角,建立包含21项误差参数的标定误差模型,以速度误差和姿态误差为观测量,建立27维Kalman滤波模型。仿真结果表明:加速度计和陀螺零偏标定误差分别优于3μg_(n)和0.00015°/h,标度因数误差优于4.5×10^(-6),安装角误差优于1″,标定时间小于30 min,实现了高精度的误差参数全标定。The calibration path design of strapdown inertial navigation system(SINS)filtering method is complex,and in⁃depth theoretical analysis is not conducted for error parameter identification.To solve the above problems,a systematic calibration method based on matrix decomposition is proposed.Fitting approach for calibration path design is introduced into the filtering method,in⁃depth theoretical analysis is performed for error parameter identifi⁃cation during stationary locations and rotation process,and a set of seven⁃position consecutive calibration path is de⁃signed.The installation error angle is decomposed into non⁃orthogonal angle and misalignment angle based on matrix decomposition.The established calibration model contains 21 error parameters.The velocity error and attitude error are taken as the measurement to establish the 27⁃dimensional Kalman filter model.The simulation results show that the accelerometer bias error is less than 3μg_(n),the gyro zero bias is less than 0.00015°/h,the scale factor error is less than 4.5×10^(-6),the installation error is less than 1″,and the calibration time is less than 30 minutes.All error parameters are calibrated with high precision.

关 键 词：捷联惯导系统 系统级标定 矩阵分解 KALMAN滤波 

分 类 号：TP212.1[自动化与计算机技术—检测技术与自动化装置] U666.1[自动化与计算机技术—控制科学与工程]