| Strapdown inertial and celestial integrated navigation system(SINS/CNS),which takes strapdown inertial navigation system as reference system and the latter as assistant to obtain high-precision navigation and positioning parameters,has high autonomy and high reliability.It is the preferred navigation system for new generation space aircrafts such as satellite,deep space detector and ballistic missile.And it is necessary to reduce the errors and optimize the information fusion algorithm in order to achieve high-precision navigation and positioning.In this context,this paper analyzes and studies the errors of the inertial devices in the active phase,optimizes the key algorithms of SINS/CNS and designs a SINS/RF-CNS integrated navigation system which can effectively estimate the accelerometer offset,so as to provide theoretical reference for SINS/CNS integrated navigation applied in the space environment.Firstly,starting from the strapdown inertial navigation system and the celestial navigation system,the basic working principle and solution method are discussed.Taking the launch point inertial system as the navigation coordinate system,the integrated navigation system mode is analyzed,and the SINS/CNS integrated navigation state model is established.Aiming at the high dynamic environment of the boost phase,variations of any error sources will greatly affect the navigation accuracy.The traditional error model cannot fully reflect its error condition.In this paper,a multi-source error model for the inertial components in the high dynamic environment of the boost phase is established by analyzing the complex situation of the forces on the body and the changes of the environment,considering the factors such as sensor error,installation error,scale factor error,lever arm error,accelerometer quadratic error,coning motion and line vibration.Through the construction of multi-source error simulation experiment analysis platform,simulation experiments are carried out,the proportion of multisource error in the system error and the coupling of different error sources in the system are discussed.Also,the influence of different error sources on the lateral drift and longitudinal drift of the missile body is analyzed.Then,the traditional SINS/CNS integrated navigation system based on attitude can only estimate the gyro drift effectively,but it cannot do anything about the accelerometer bias,which affects the missile positioning accuracy.In this paper,an improved SINS/RF-CNS integrated navigation mode is proposed based on the indirect sensitive horizon mode of star refraction.Firstly,through triangle algorithm,continuous star map simulations and star map matching recognition are realized.Besides,the relationship between refraction apparent height and missile position is deduced,and then,a new nonlinear integrated navigation model is established with the kinematic constraint of missile.Secondly,corresponding observation noise model is studied,then the influence of the refracted light's refraction height and refraction angle error on the measurement information is analyzed through experiments,so as to improve the positioning accuracy and the system robustness.Finally,due to the limitation of engineering experiment conditions,the system's experimental simulation platform is built,and the integrated navigation simulation experiments are completed with ballistic missile.The system positioning effects achieved by traditional integrated navigation mode and SINS/RF-CNS integrated navigation mode based on EKF and UKF algorithm under different initial misalignment errors conditions are compared and analyzed.The experimental results confirm that the proposed SINS/RF-CNS integrated navigation method has a significant effect on estimating the accelerometer offset and improving the positioning accuracy.The three-axis position error is increased by 84.27%,89.53% and 85.02% compared with the traditional method respectively. |
PDF Full Text Request