| Micro inertial measurement unit(MIMU)consists of three micro gyroscopes and three micro accelerometers.With the advantages of small volume,cheap cost,low power consumption,easy to mass production and so on,it has been widely used in civil and commercial inertial navigation systems.Low-cost MIMU has low measurement accuracy,which limits its further application.In this paper,the research of micro inertial sensor array based on information fusion technology is carried out in this context,including the following aspects:1.The error model of MEMS inertial sensor is studied,and the random error of the gyroscope and the accelerometer is identified by Allan variance modeling method.A signal generator is built by using Simulink,and the influence of deterministic errors on Allan analysis of variance was studied.2.The application of Kalman filter in micro inertial sensor array is studied,the Kalman filter is designed and the effectiveness of the filter is verified by Simulink simulation.3.An array consisting of three MIMUs and a corresponding data acquisition system are designed and built,a prototype of the micro inertial sensor array is produced,a host computer is written,and a functional test of the prototype is performed.4.The calibration method of the array is studied.The MIMU is calibrated through turntable experiments and temperature experiments to complete the error compensation.5.The static experiment,uniform rate experiment and swing experiment of the array prototype are carried out with turntable.The Kalman filter is state-amplified for dynamic conditions,and a fade-out factor is introduced.The experimental results show that the standard deviation of the signal decreases from 0.3752 ° / s before filtering to 0.0684 ° / s after filtering under the condition of constant velocity.In the swing experiment,the standard deviation of the signal decreases from 0.3946 ° / s before filtering to 0.1814 ° / s after filtering,and the filtering effect is better than the weighted average method.. |
PDF Full Text Request