Research On Initial Alignment Algorithm Under Large Misalignment Angles

Nowadays,with the rapid development of science and technology,the demand for high-precision navigation technology in various fields of social production is increasing.Strapdown inertial navigation has been widely used as an independence navigation method,as the important premise and key technology of inertial navigation,initial alignment technology are the crux to ensuring the proper functioning of the inertial navigation system.In recent years,the carrier needs higher navigation accuracy and faster starting speed,the index corresponding to the initial alignment is the accuracy and speed of the alignment.This paper uses this as the research background to study the one-step initial alignment under large misalignment angles,nonlinear filtering and inertial frame fine alignment methods.First of all,this article briefly introduces the commonly used reference frames and coordinate transformation methods in inertial navigation,derives the update algorithm and error equation of strapdown inertial navigation(SINS),and then introduces the basic principles and equations of the Kalman filter algorithm;The integrated navigation method and discussed two information fusion methods:direct approach and indirect approach and analyzed their respective advantages and disadvantages,laying a theoretical foundation for the following research on the combined alignment algorithm with large misalignment angles.Secondly,this article explains the basic idea and update process of the Unscented Kalman Filter(UKF)accordingly,and introduces the Iterative Unscented Kalman Filter(IUKF)derived from the Gauss-Newton iterative idea.Then the SINS equation under the large misalignment angle is derived and simplified to different degrees in different applications,and then the SINS large misalignment angle alignment model is decomposed into the nonlinear part and the linear part,and the UKF/KF combined filter is used to reduce the calculation amount of UKF and improve the real-time performance of the solution.Combined with the simulation,the iterative method of IUKF is discussed in detail,and the IUKF/KF algorithm applied to the situation where the three angles are all large misalignment angles is proposed,and the simulation verifies that IUKF/KF has a faster convergence rate than using UKF alone.And it has more advantages in accuracy.Then,according to the Euler angle differential equation and the specific force equation,a direct filtering alignment model with a large misalignment angle in the geographic reference frames is established,the linear and nonlinear measurement equations are established respectively according to the GPS and odomter measurement.Through IUKF To improve the problem of insufficient utilization of measurement information caused by the nonlinearity of the measurement equation when the odometer is used for measurement,and the feasibility of direct method alignment is verified by experiments on vehicle-mounted moving bases.Subsequently,the problem of singularity in the differential equation when Euler angles expresses attitude is studied.The two sets of positive and negative Euler angles are used to establish different system equations and switch in real-time to avoid singularities.When the roll angle is around 90 degrees,the optimal estimation of the system state can still be ensured.Finally,in order to extend the idea of the solidification of reference frames of the inertial frame coarse alignment to the fine alignment,the integrated inertial system specific force and the inertial system velocity are measured as the quantity,and two indirect inertial system fine alignment methods are studied.Derived the direct method of inertial system fine alignment using simplified UKF,and finally verified the advantages and disadvantages of these three inertial system fine alignment methods and traditional precision alignment methods through the sea trial data of the shaking base and the vehicle-mounted data of the moving base.