Motion Traits Of Ballistic Missilies And Methods Of Their Measurements Fusion

Tracking precisions of ballistic missiles are of important concerns to shooting ranges, which include real time tracking precisions and post processing precisions. There are two approaches to improve the precision of missile tracking, which are performances enhanced of measurement equipments and better methods of data processing. Aim at better methods of data processing, this paper studies methods to improve the precision of missiles tracking by two approaches: traits of trajectories and their applications in data fusion processing.Tracking models are usually built by two ways: the physical properties of missiles and motion properties statistics, which both utilize prior information of trajectories. This essay firstly analyses how forces act on missiles during their flights, which consist of boost phase, coast phase and reentry phase. And then dynamic models, fit for different flight phase, are built based on the analysis. Especially for the reentry phase, models of two different situations, maneuvering and nonmaneuvering, are researched. Then kinematic models are studied based on the motion properties statistics. The simulation indicates the real time traking method based on the CS tracking model (current statistics model) is better during boast phase.For the ballistic coefficient refects the property of missile deceleration during reentry phase, without considering effect of gravity, then the variation of ballistic coefficient indicats the motion of missiles during their reentry flights. Different models of ballistic coefficient are studied and the properties of these models are comparied by simulation. Also, the Cramer-Rao bound is introdued to compare the filtering effect of EKF and UKF to study the real time data processing, and a general conclusion is given.For the part of data fusion, the studies are progressed by way of real time data processing and post processing. For the former, steps of real time data fusion are studied; aim at current measurement equipment, in allusion to the four real data fusion models, the simulation indicates the method, based on centralized fusion of the measurement data, is the best one, and the result offers a conclusion to the practical data processing. For the latter one, post processing, for the method of EMBET based on the spline, a new optimizing method of nodes adaptive choosing is introduced and the simulation testifies the method; applying the dynamic models to the post fusion, a method of data fusion based on the dynamic model is got.The researches in this paper provide basic theories and criterions in applications.