Research On Transfer Alignment Of Large Misalignment Angle Based On MEMS Strapdown Inertial Navigation System

In order to ensure autonomy and consider the cost,volume and other factors,the strapdown inertial navigation system(SINS)based on MEMS is generally adopted in the navigation system for precise-guided weapons nowadays,and the transfer alignment scheme is mostly taken in the initial alignment of the SINS.For the transfer alignment in the MEMS-SINS,the following two problems are mainly faced in the practical application.Firstly,the error characteristic of the MEMS inertial components is complex and greatly affected by the environment temperature.Secondly,the system error characteristic is nonlinear.These problems will affect the initial alignment accuracy of MEMS-SINS.In this paper,the research is mainly on the transfer alignment of MEMS-SINS with large misalignment.For the problem of the low accuracy and long alignment time of the MEMS-SINS,the temperature error characteristic of the MEMS inertial components was studied,the error model of transfer alignment was established,and the transfer alignment matching method was analyzed.An improved UKF filtering algorithm was proposed to solve the gaussian white noise hypothesis and the problem of heavy computation in the standard UKF filtering algorithm.The main contents are as follows:(1)The temperature error characteristic of MEMS inertial components was investigated,and the parameters of the error model were determined after the establishment of the temperature error model of the MEMS inertial components.Then the temperature error model was used to compensate the error of the MEMS inertial components to reduce the temperature drift error of MEMS inertial components.(2)The nonlinear error characteristic of the system was studied and the error model of large misalignment was established.The methods of velocity matching transfer alignment,attitude matching transfer alignment and velocity plus attitude matching transfer alignment were analyzed comprehensively before the summary on the characteristic of the matching methods.The matching method of velocity plus attitude is finally determined based on the characteristic of transfer alignment of MEMS-SINS.The influence of lever arm effect and flexural deformation on the accuracy of transfer alignment was studied,the lever arm effect error equation and deflection deformation error equation were established,and the compensation scheme was designed.(3)In term of the hypothesis of gaussian white noise and heavy computation in standard UKF,an improved UKF filter algorithm was proposed based on the standard UKF,which introduced the Huber cost function and spherical simplex transformation that would resolve the problem of colored noise existing in the system noise and the heavy computation.The final result verifies the practicability and effectiveness of the algorithm.(4)The improved filtering algorithm was verified through the car test,which shows that the transfer alignment accuracy and rapidity of improved UKF filter algorithm satisfied the demands of airborne precision-guided weapon,and the estimation accuracy of attitude angle is greatly increased than the standard UKF filtering algorithm.