Research On Integrated Navigation Technology For Low-cost Miniature Loitering Missile

With the development of precision guided weapon system,loitering missile has become one of the key research areas,which has extensive application value and far-reaching significance.With the technical background of integrated navigation system for low-cost miniature loitering missile,the scheme which integrates MEMS-IMU,GPS,magnetic compass and altimeter is proposed.This dissertation will focus on several related key techniques for integrated navigation system,including sensor error analysis and compensation,initial alignment,and multi-sensor data fusion(MSDF).The main work is described as follows:1.The measurement error characteristics of the sensors used on loitering missile are researched.As for the systematic error of MEMS inertial sensors,the error model is established and corresponding parameters are calibrated on rotating platform using six-position method and rate test,as well as the installation error and lever arm of the IMU.As for the stochastic error of MEMS inertial sensors,the major error terms and parameters are identified off-line using Allan variance,and eliminated online using adaptive filter algorithm.As for magnetic compass,the experiment of azimuth measurement is carried out to compensate its ellipse error.2.The error model of integrated navigation system is established,and initial alignment method for large azimuth misalignment angle is studied.Under the condition of static base,coarse alignment is accomplished with the aid of magnetic compass.And the states of the filter,which are not observable,are removed to reduce computation burden.Under the condition of moving base,the trigonometric variables of the azimuth,rather than the azimuth,are used as estimated states to establish the navigation error model.Refer to the navigation information of position provided by GPS,precise alignment for large azimuth misalignment angle can be accomplished.Aiming at the observability analysis of inertial alignment system with external information aided on moving base,a novel approach that starts from the basic observability definition is proposed to analyze original nonlinear state equations directly.Different with traditional linearization-based observability studies,this kind of analysis approach is straightforward and comprehensive,with improvement of accuracy,computation and physical meanings.3.The information fusion technology of integrated navigation system is studied in aspects of the structure,algorithm and model.The multistage information fusion structure of the attitude,velocity and position is determined.Algorithms of Sage-Husa adaptive extended Kalman filter and strong tracking Kalman filter are combined and improved,used to estimate system states variance.On third level structure of vertical position and velocity information fusion,federated filter and fault diagnosis are added to enhance the fault-tolerant ability of the system.The filter models of each level of navigation information fusion structure are established.Finally,covariance simulations are carried out,and numerical results alanysis are given.