MIMU gyroscope random drift forecasting method based on ARMA and BPNN combination model

Abstract

The invention discloses a MIMU gyroscope random drift forecasting method based on an ARMA and BPNN combination model. The method mainly includes the steps that 1, random drift data of a single-axle micro gyroscope is collected in the static state, the stationarity of an original data sequence of a gyroscope is judged by observing graphic characteristics of an autocorrelation parameter ACF and a partial correlation parameter PACF, an ADF unit root is checked, and stationary data is obtained after a trend term of the sequence is removed through differences; 2, the stationary sequence of the gyroscope is subjected to model order determination with the AIC minimum rule method, and an ARMA model of a gyroscope drift stationary sequence is established; 3, a random drift sequence of the gyroscope is modeled through the combination model, a training set and a test set are selected from error data of the ARMA model in the step 2, and a BP neural network forecasting model is established, and the structure of a BP neural network is set; 4, sample data of the established BP neural network is trained, and the forecasting result of gyroscope random drift is preserved.

Description

technical field [0001] The invention relates to the random drift prediction of MIMU (Micro Inertial Measurement Unit) in a deep integrated navigation system, in particular to a gyroscope random drift error prediction and fitting method. Background technique [0002] With the continuous improvement of integrated circuit silicon semiconductor manufacturing technology, micro-machine manufacturing technology has made great progress, and MEMS (Micro-Electro-Mechanical System) inertial sensors have emerged. At the same time, it has the characteristics of small size, light weight, high reliability, impact resistance, easy installation, mass production, and low cost. It is integrated with microelectronic processing circuits to achieve mechatronics. In particular, with the continuous improvement of the precision of MEMS inertial sensors, in the navigation system equipment in the military field, the strapdown inertial navigation system using micro-mechanical inertial instruments as in...

AI Technical Summary

Problems solved by technology

[0004] In the existing gyroscope random error modeling, common methods to reduce gyroscope random drift are Allan variance method, wavelet analysis method and Kalman filter method. Although Allan variance method is easy to calculate and easy to separate errors, it is easy to leak power and Quantitative expression method is single

The wavelet analysis method can decompose the signal in any detail, and is suitable for analyzing and processing non-stationary signals. However, if the number of wavelet layers is not selected properly, useful signals will be filtered out.

The Kalman filter method can be used for the estimation of time-varying, non-stationary and multidimensional signals, but the noise must be assumed to be a Gaussian process whose statistical properties are known

Images