Research On MEMS Gyroscope Array Technology Based On Ulman Filtering And Data Fusion

Because the micro-electro-mechanical(MEMS)gyroscope has the characteristics of small size,low price,and easy mass production,its application prospects are broad.At present,its accuracy is not high enough due to the limitation of technology and other conditions.It is mainly used in low-end fields such as consumer grade,and it is difficult to apply in high-precision fields.Therefore,how to reduce the noise of the gyroscope and improve its accuracy has become a hot issue in this field.In recent years,breaking through foreign technical blockades and independently developing high-precision MEMS gyroscopes has more practical significance.Aiming at the problem of insufficient accuracy of MEMS gyroscopes,based on the existing performance tests,the thesis first analyzes the research and application background of virtual gyroscopes,and the current research status at home and abroad.Secondly,the error source,error identification,and random error modeling methods of MEMS gyroscopes are discussed.Finally,this paper combines multiple unrelated and same model gyroscopes to form a gyroscope array to form a virtual gyroscope,and proposes a multi-stage filter The algorithm improves the output accuracy of the virtual gyroscope and reduces its noise.The main innovations of this article are as follows:1)An improved Kalman filter algorithm is proposed.In practical engineering applications,the variance of the output data of the gyroscope is fluctuating,so on the basis of the classical Kalman filter,this paper obtains the functional relationship between the angular velocity and the variance through experiments,and takes it as the Kalman filter.Filtering the value function of the R value,an improved Kalman filter is obtained.The experimental results show that four MEMS gyroscopes with a standard deviation of 1σ of 0.005°/s have reduced the standard deviation of a single gyroscope by 4 times after the improved Kalman filter,which is 48%of the standard deviation of a single gyroscope after the Kalman filter.2)A distributed multi-stage filtering method is proposed.First,the improved Kalman filter is used to filter each component gyroscope to obtain a state estimation value;second,the maximum posterior estimation is used to derive the inverse smoothing estimate,and each component gyroscope is subjected to two-stage filtering to obtain the second state.Second state estimation value;finally,the principle of minimum error covariance trace is adopted to obtain the weight value of each component gyroscope,and its state estimation value is weighted and averaged,so that the output noise of the virtual gyroscope has been greatly reduced.The experimental results show that the four MEMS gyroscopes with a bias stability of 1.65°/h have improved the bias stability by 550 times after using distributed multi-stage filtering.After a large number of experimental verifications,the virtual gyroscope designed in this paper has a significant improvement in accuracy compared with a single gyroscope,can effectively replace the traditional MEMS gyroscope,and has greater market prospects and application value.