Research On MEMS Gyroscope Error Analysis Technology Based On Improved SHAKF Algorithm

MEMS gyroscope can provide angular velocity,azimuth,level,position and other information,and has the advantages of small size,low power consumption and good environmental adaptability.It has important applications in industrial equipment,consumer electronics and other fields.With the increasing demand of MEMS gyroscope,new requirements are put forward for its accuracy.In order to improve the accuracy of MEMS gyroscope without hardware breakthrough,it is beyond the performance limit of a single MEMS gyroscope.In this paper,virtual gyroscope technology and improved Saga-Husa adaptive Kalman filter algorithm are combined to improve the accuracy of MEMS gyroscope.By designing a virtual gyroscope system based on four gyroscope arrays,four gyroscopes are used to simultaneously measure the same angular rate signal,and the best estimation signal is obtained by data fusion.Multiple low-cost MEMS gyroscopes form a high-precision gyroscope array,which can not only reduce the gyroscope cost in the field of inertial navigation,but also achieve higher accuracy than a single gyroscope.Aiming at the problem that the filtering parameters of the traditional Kalman filter algorithm are fixed and easy to lead to filtering divergence,the Sage-Husa adaptive Kalman filter algorithm is improved based on covariance matching technology,and an improved Sage-Husa adaptive Kalman filter algorithm is proposed.The proposed algorithm can estimate and correct the noise statistical characteristics in real time.The angular random walk of the gyroscope array is reduced from 0.39 °/ h^1/2 to 0.044 °/ h^1/2,and the bias instability is reduced from 71.11 °/h to 8.27 °/h.The random noise of MEMS gyroscope caused by external factors such as environment is complex and irregular,which greatly affects the accuracy of MEMS gyroscope.According to the characteristics of random error of MEMS gyroscope,a random error model is established.Using the fully overlapped Allan variance method with higher estimation accuracy,the random errors of single gyro and gyro array before and after filtering are identified and analyzed.It provides parameter support for verifying the effectiveness of the filtering algorithm.When MEMS gyroscope works in harsh environments such as vibration,Allan variance cannot analyze the dynamic characteristics of its noise.Aiming at the problem of fixed dynamic Allan variance window,a PID-DAVAR adaptive algorithm is proposed.The length of the truncation window is adaptively adjusted according to the dynamic characteristics of the gyro output signal.The average tracking error of noise coefficient of angle random walk,bias instability and rate random walk is about 10 %,and the minimum error is about 4 %.It only takes 8.65 s to calculate and draw the DAVAR three-dimensional map of 60,000 data.The PID-DAVAR adaptive algorithm not only ensures the variance confidence,but also shortens the data processing time without losing the signal characteristics.It provides an effective method for the analysis of dynamic noise of MEMS gyroscope.In this paper,a virtual gyroscope system based on MEMS four gyroscope array is designed,and the gyroscope data acquisition,analysis and processing are completed.The gyro error is identified and analyzed by completely overlapping Allan variance.PID-DAVAR adaptive algorithm is proposed.The improved Sage-Husa adaptive Kalman filter algorithm is used to filter the gyro data.The static and dynamic experimental results prove the feasibility of the system in improving the accuracy of MEMS gyroscope.