Research Of Gyroscope Measurement Accuracy Based On Kalman Filter

The photoelectric stabilized platform for aviation realizes the automatic tracking and recognizing to the target by the image detection device,and is widely used in the military field and civil environment such as public security,environmental monitoring.The gyroscope is the core sensitive sensor of the photoelectric stabilized platform for aviation.It is mounted on a moving carrier,measure the velocity changes to isolate the carrier disturbance and ensure the platform to maintain stability as much as possible,so as to make sure the optic axis of the device that installed on the photoelectric stabilized platform for aviation has a stable point.Due to the adverse impact of external environment and other reasons such as the gyroscope's structure errors,the gyroscope in the platform can introduce the random drift and noise in measurements,resulting in the output torque of the DC drive motor can not completely drive the platform frame move accurately,further affect the optic axis stability and accuracy.Therefore,in order to improve the LOS stabilization accuracy of photoelectric stabilized platform for aviation,the gyro sensor output signal must be effective filtering.Kalman filter is the optimal estimation theory which describes the dynamic system according to the state equation.The linear minimum variance is regarded as the estimation criteria,and the prediction and correction methods are used to estimate the state of system.Kalman filter is widely used because of its advantage of real-time recursion,small storage space and ease of use.Combined with the actual engineering needs,this paper's target is military used two-axis four-gimbal photoelectric stabilized platform for aviation.The strong tracking Kalman filter algorithm is used to filter the output signal of gyroscope sensor,to improve the precision of the velocity loop control system of the platform.The main contents of this paper are as follows:(1)Firstly,according to the working principle of the two-axis and four-gimbal photoelectric stabilized platform for aviation,the disturbance factors existing in the system are synthetically analyzed.Then discuss the adverse effect of the measurement noise of the gyroscope on the stability of the optic axis.(2)In order to effectively filter the output signal of the gyro sensor,a mathematical model is established for the gyro signal.The modeling method adopted in this paper is time series method,which can predict the internal and external influence of the system and predict the future situation of the system according to the output data of the system.Considering the complexity of the algorithm and the computational load in the practical application,the AR(1)model is the mathematical model of the gyro sensor,which provides the first condition for the application of Kalman filter.(3)Based on the mathematical model of the gyro sensor,the state equation and the measurement equation of Kalman filter are constructed,and the measurement noise of the gyro signal is suppressed by Kalman filter technique.At the same time,a strong tracking algorithm is introduced to adjust the state predictive variance in Kalman filter in real time.Forcing the output error sequence to remain uncorrelated,the strong tracking Kalman filter is constructed to enhance the robustness to the system model uncertainty.(4)The strong tracking Kalman filter proposed in this paper is compared with the second order Butterworth filter which is commonly used in engineering to verify the filtering effect and robustness.The experimental results show that the amplitude and the variance of the output signal of the strong tracking Kalman filter are obviously better than the output signal of the second order Butterworth filter.And the strong tracking Kalman filter can optimize the dynamic performance of the control system,guarantee the signal's real-time performance.Compared with the standard Kalman filter,the strong tracking Kalman filter proposed in this paper has better robustness and enhances the stability of the control system.