Research On MEMS Gyro's Error Compensation Method Based On Micro Inertial Navigation System

Micro inertial navigation system is made up of inertial device(micro silicon gyroscope,micro silicon accelerometer),computer and so on,it has many advantages,such as good autonomy,good concealment and can provide a variety of navigation parameters.But with the accumulation of time,its error was becoming more and bigger.The error is divided into the system-level error and the inertial device-level error,the former can be compensated by a certain method;However the micro inertial devices is very susceptible,the precision is very low,it was serious influence the outcome of the navigation,and it was the main source of inertial navigation system error,so for micro inertial devices error compensation has very important meaning.This thesis mainly to analyze the error of MEMS gyro and modeling,combining with the practical application environment and using the effective method to modeling and error compensation,to reduce the error of inertial navigation system and improve the positioning precision plays an important role.MEMS(Micro Electro Mechanical Systems,referred to as MEMS)gyroscope compared with the traditional gyroscope,it has many advantages such as small volume,light weight,easy integration and so on.In weapons navigation,car navigation and other military and civilian fields have broad application prospects.However,due to its accuracy is influence of manufacturing processes and work environment,greatly limited the wide application of MEMS gyroscope,so for research MEMS inertial devices has become the focus of national research project.This paper mainly for MEMS gyro error analysis and compensation studies,the micro mechanical gyro will widely use in high precision navigation system become possible.Firstly,the article was briefly introduction the inertial navigation system,the strapdown inertial navigation system as main research object,and deeply researches the principles of inertial navigation systems,components and sources of error.Secondly,studied the gyro that is the important component of the inertial navigation system,in understanding the basis of the micromachined gyroscope structure and working principle,and detailed study the MEMS gyroscopes' s error characteristics.MEMS gyroscope's error mainly consists of two key components: random error and deterministic error that including composed of zero error and installation error,deterministic error can be compensated by experimental method;Random error gradually increasing with time and no rules to follow,and very difficult to establish accurate mathematical model to describe it,so the random error is the main factor influencingthe MEMS gyro precision.This paper based on the MEMS gyro's sampled data characteristics,using the wavelet analysis method to analysis and processing output data of gyro,and the nonlinear modeling's method of wavelet neural network(Wavelet Neural Network,WNN)is applied to MEMS gyroscope random error modeling,the simulation results show that the this method achieved good results in the MEMS gyroscope random error modeling and compensation,reduce the impact on the accuracy of the MEMS gyroscope.Based on the above research,this paper briefly studied the micro-inertial navigation system based on MEMS,by using the method of this paper to modeling and compensation the random error of the selected MEMS gyroscope,and this method is introduced to the inertial navigation system.The experiment results show that the wavelet neural network modeling method has good application value on improving the MEMS gyro performance.