Research On Error Suppression And Compensation Methods Of Strapdown Inertial Navigation System

Due to its advantages of autonomy,continuous and comprehensive output information,strapdown inertial navigation system(SINS)is widely used in land,sea and air navigation fields.However,gyro errors,accelerometer error,initial alignment error and other error sources lead to Schuler,Earth and Foucault periodic oscillation errors of SINS.Moreover,the random error component of gyro will produce the cumulative error component whose root mean square increases with the square root of time,and the cumulative error component reduces greatly the real-time navigation precision of SINS.In view of the above problems,this paper carried out research works on error suppression and compensation methods of SINS.In order to overcome shortcomings of the traditional damping technology,such as large overshoot error,long error convergence time and so on,the adaptive integrated navigation technology,information fusion technology and oscillation error suppression technology are discussed deeply based on Kalman filtering technology.At the same time,the error compensation method of undamped SINS is presented to improve the navigation precision of SINS.The mainly work is as follows:Firstly,establish the relevant models of SINS and integrated navigation system,such as system coordinate systems commonly used by SINS,mechanization equations of SINS,error equations of SINS,mathematical model of integrated navigation system and so on.In order to solve the problem that the filtering accuracy decreases or even diverges,which caused by the unknown statistical characteristics of noise,the adaptive Kalman filtering algorithm of integrated navigation system is deeply studied.To improve the traditional innovation adaptive Kalman filtering algorithm,the gradient detection function of the measurement noise variance matrix is designed to detect the actual change of the statistical characteristics of the noise in real time.According to detection results of the gradient function,the sampling interval width of the innovation detection sequence is adjusted in real time through the adaptive sliding window function,and the window width changes adaptively.Simulation results show that the new adaptive innovation integrated navigation algorithm based on sliding window can change adaptively the window width according to the actual situationcan to solve the difficulty of selecting window width and improve the navigation precision of the integrated navigation system effectively.The tracking precision and tracking sensitivity of the adaptive filtering algorithm are both considered at the same time.Mesnwhile,the computation is relatively small and the realtime is high.Secondly,the traditional damping technique to suppress oscillation errors of SINS is discussed and analyzed in this paper.Aiming at overcoming its shortcomings including large overshoot and long error convergence time,a complete damping oscillation error suppression method based on Kalman filter is proposed.Using the measurement information of accelerometer and the calculation parameters of SINS,the projection of velocity change rate on the geographical coordinate system is derived,and it is integrated in a certain fixed period of time to separate carrier velocity that is compensated by external reference velocity.The compensated integral results are used as Kalman filtering observations,and the Kalman filtering observation equation is derived.Kalman filter is used to estimate the error state of SINS,and oscillation errors are compensated by output correction or feedback correction.Subsequently,the steady-state error of the complete damping method based on Kalman filter is derived and it is compared with the steady-state error of the traditional damping technique.Simulation and test results show that the proposed method can suppress Schuler oscillation errors,Earth oscillation errors,Foucault oscillation errors simultaneously.Further more,compared with the traditional damping technology,the time of error convergence is greatly shortened,and the overshoot error caused by the damping state switching is significantly reduced.Thirdly,the error of integrated navigation algorithm with velocity error as an observation is discussed and analyzed.In order to further shorten the error convergence time and improve the navigation accuracy of the complete damping method based on Kalman filter,oscillation error suppression method of SINS based on dual filters is proposed.The serial structure of dual filters is designed to make full use of the velocity smoothing effect and the error convergence speed of integrated navigation algorithm.Based on the estimation error covariance matrix,the information of two filters is fused,and the state estimation value correction function is designed to suppress the influence of the external velocity constant error on the system error.Simulation results show that the proposed method can effectively shorten the error convergence time,restrain the influence of the external velocity constant error on the system,and improve the navigation accuracy.Lastly,the error propagation mechanism of SINS is deeply analyzed,and the error of undamped SINS is analyzed.To solve the problem that the oscillation error can not be compensated when the carrier is in the state of non-uniform motion or external reference information is unavailable,an oscillation error compensation method for undamped SINS is proposed.Benefiting from that SINS can execute multiple parallel navigation algorithm at the same time,the periodic oscillation signals whose phase difference is ? can be counteracted by the average cancellation principle of periodic oscillation signals.Periodic oscillation errors are compensated by means of predicted time series and delayed alignment.Simulation and test results show that the proposed method can compensate periodic oscillation errors of undamped SINS to effectively improve the accuracy of SINS while retaining the autonomy of SINS.