Research On The MIMU Error Analysis And System Level Calibration Method

Nowadays,Micro-Electro-Mechanical System(MEMS)inertial sensors with small size,low cost and high integration are developing rapidly.The Micro-Inertial Measurement Unit(MIMU)based on MEMS gyro and MEMS accelerometer has a broad prospect as the core component of inertial navigation system.At present,MEMS inertial sensors with small volume,low cost and high integration are developing rapidly.The micro-inertial measurement combined unit(MIMU)based on MEMS gyro and MEMS accelerometer,as the core component of inertial navigation system,has a broad application prospect.However,due to the limitation of manufacturing process,its accuracy is generally not high.It is necessary to improve the accuracy of MIMU by means of software calibration and compensation without changing the hardware of the system.According to the different observation quantity,the calibration methods can be divided into discrete calibration method and system-level calibration method.However,the existing discrete calibration methods rely too much on calibration equipment,and the steps are complicated and the process is tedious in the actual engineering application.In the system-level calibration method,the attitude and rotation information are provided by the low precision turntable,and the MIMU output information of the calibration position and rotation process is introduced into the navigation calculation,and the error terms of MIMU are identified by observing the navigation results.Therefore,it is of great practical significance to develop a system-level calibration method for MIMU errors that is fast and convenient and independent of high-precision testing equipment.Firstly,the working principle and error mechanism of MIMU are analyzed,and reasonable error models are established for constant error and random error respectively.The Allan variance was used to analyze the MIMU random error.Secondly,a system-level calibration scheme of MIMU was designed,and the basic principle and algorithm were deduced theoretically.The mathematical model of navigation speed error and MIMU constant error was established with the speed error as the observation quantity,and each error item of MIMU was identified.A rotation arrangement scheme for system level error calibration of MEMS gyro and accelerometer is designed.Finally,the MIMU system level calibration software is designed and developed,and the errors of MIMU are calibrated and compensated online.Through error iteration compensation experiments,the accuracy of MIMU is effectively improved,and the accuracy of the calibration method and the reliability of the software are verified.The results show that the MIMU system level calibration method is a low cost,fast and reliable MIMU calibration scheme.