The Study On Track Estimation And Threat Assessment In Multi-source Information Fusion System

Multi-source information fusion is a multi-level data processing process,including detection,correlation,combination and estimation of multi-source data.It can be used to improve the accuracy of the target state and the identity estimation,and then form a comprehensive evaluation of the battlefield situation.In recent years,the combat equipment has been upgraded constantly,and all kinds of camouflage stealth technology and interference technology have been widely used in the battlefield.This leads to the failure of target tracking and recognition based on only one single sensor.It is urgent to develop a fast and effective information fusion system.In this paper,based on the characteristics and difficulties of maneuvering target tracking and evaluation,a method of target track estimation and threat assessment based on multi-source information fusion is proposed.The research results of this paper can realize the high precision adaptive tracking of maneuvering targets and effectively transform the target track data into decision level information,which can provide a basis for the development of attack plans.In order to meet the requirements of actual combat,the architecture of information fusion system is established based on the analysis of the function model and architecture of the fusion system.Then,five kinds of information sources which are commonly used in information fusion system are analyzed.According to the application scenario,the appropriate information sources are selected.Currently used filtering methods are compared and analyzed.An improved interactive multiple model algorithm is proposed,which combines the advantages of the adaptive tracking of the interacting multiple model algorithm and the good filtering performance of Unscented Kalman filter for nonlinear system.Then,combined with the motion characteristics of the maneuvering target,the motion model set of interacting multiple model algorithm is established.The simulation results show that the improved interactive multiple model algorithm can effectively improve the tracking accuracy of maneuvering target,and the model set can be well fitted with the moving characteristics of the air target.The advantages and disadvantages of the classical distributed track fusion algorithm and its application conditions are analyzed.The hierarchical track fusion algorithm is selected because of its small computation and high fusion accuracy.In order to solve the problem that the hierarchical fusion algorithm cannot be used to estimate the results of interacting multiple models,a hierarchical fusion algorithm based on interacting multiple models is proposed by constructing the whole system priori statistics.The performance of the proposed algorithm is verified by simulation experiments,and the fusion results are compared under two modes of dual radar,dual radar and one infrared.Simulation results show that the fusion algorithm can effectively improve the estimation accuracy,and the use of heterogeneous sensors can play the advantages of different types of sensors,so that the estimation accuracy can be further improved.On the basis of acquiring the target track,the target threat assessment method is studied.As the traditional target threat assessment is always subjective and the weighting method is single,an aerial targets threat assessment method based on analytic hierarchy process(AHP)and entropy weight method is proposed.The index system is established first as the basis of threat assessment.Then the weight of each index is calculated separately by using AHP and entropy weight method.Based on the weight and the attribute matrix,the combination weighting model under the principle of minimum total deviation is constructed.The model is applied to multiple attribute decision making method and the comprehensive threat assessment of each aerial target is obtained.The example shows that the method gives attention to both the expert experience and the objective characteristic of battlefield data,the weight is more reasonable and the threat assessment is more conform to reality.