Automatic switching and denoising MEMS-Gyro signal for enhanced dynamic performance

The state model and measurement accuracy of the system can be changed according to the dynamic condition, and different filtering methods are proposed according to the different rate of the signal. It is proposed to build switching system, combing such methods as time series analysis model, the least square, and adaptive moving average algorithm together to form a switching noise reduction method. Adaptive Kalman filter based on state correction is studied to improve the filtering accuracy of the system and tracking performance. The method is finally testified by simulation results, and it shows that the system noise can be reduced, the system changes be tracked dynamically, and the dynamic noise reduction effect of the system be improved.