Technical Research Of Inertial Navigation System Based On Redundant MEMS Gyroscope

As an autonomous navigation system,inertial navigation system(INS)is widely used in aerospace,military weapons and civil fields because of its strong autonomy,good concealment and wide working environment.The increasingly complex requirements in various fields make the inertial navigation system develop towards miniaturization,low energy consumption and high reliability.Micro-electromechanical system(MEMS)and redundancy technology provide a strong driving force for this.This thesis has carried out a series of research work around the inertial navigation system based on redundant MEMS gyroscope.The main research contents include:Firstly,the redundant configuration scheme of MEMS gyroscope is discussed.The basic principles followed by the multi-sensor redundant configuration scheme are introduced,and the reliability and navigation performance of the multi-sensor redundant configuration scheme are evaluated.By introducing the selection index function of the system's average reliability,the six gyroscopes used in this paper are finally determined Redundant configuration scheme of regular dodecahedron.Secondly,the error of redundant MEMS gyroscope is analyzed.The deterministic error and random drift error of MEMS gyroscope in redundant configuration are introduced.For the deterministic error,the error model is established,and a six-position turntable calibration method is introduced for the redundant configuration scheme adopted in this paper.For the random drift error,the five typical error sources are analyzed and identified using the Allan analysis of variance method.Next,the technology of random drift error compensation for redundant MEMS gyroscopes is studied.The basic principle and process of time series modeling are introduced,and the time series model of random drift error signal of MEMS gyroscope is established.Based on the established model,traditional Kalman Filter and Adaptive Kalman Filter are used to compensate the random drift.Aiming at the problems of the time series model,the modeling idea of the Singer model is introduced and a random drift error model based on the Singer model is established,and Adaptive Kalman filtering is used for filter compensation.Finally,the fault diagnosis technology of redundant inertial navigation system is studied.The direct comparison method and generalized likelihood ratio fault diagnosis method commonly used in redundant inertial navigation system are introduced.Aiming at the redundant configuration scheme discussed in this paper,a generalized likelihood ratio fault diagnosis method is used for simulation test,and the test results verified the effectiveness of the generalized likelihood ratio fault diagnosis method.On the basis of fault diagnosis,the reconstruction scheme of redundant inertial navigation system is introduced.