Research On Key Technologies Of Aerial Target Tracking And Prediction

The environment of modern air battlefield is complex and changeable,all kinds of advanced aircraft and sensor equipment are constantly updated and iterated,electronic jamming means emerge in endlessly,and the problems of strong maneuver and high clutter in target tracking are increasingly prominent,which puts forward more strict requirements for the ability and accuracy of air target tracking prediction,and has high research value in practical engineering application.In this paper,based on the existing target tracking prediction algorithm,the dynamic model,filter and interactive algorithm of maneuvering target are deeply studied.Aiming at the model mismatch problem in the conventional multiple model algorithm,a multi axis model set matching algorithm is proposed to adapt to the different maneuvering characteristics of the target.Secondly,aiming at the corresponding problem between the filter and the target maneuvering situation in the existing filtering algorithm,a multi filter selection prediction algorithm is proposed to improve the trajectory tracking prediction accuracy of targets with different maneuverability.The achievements of this paper mainly include the following aspects:(1)In order to solve the problem that the model set is fixed and the number of models is limited,which can not cover all the maneuvering states of maneuvering targets in three-dimensional space,a multi axis model set matching(MMSM)algorithm is proposed.The algorithm considers the inconsistency of the maneuvering state of the air target in different directions,decomposes the maneuvering state of the target into three coordinate axes,and matches the model on each coordinate axis respectively,so as to improve the model matching rate of the whole algorithm.The simulation results show that the model mismatch problem of the multi axis model set matching algorithm is obviously improved and the tracking effect is better than that of the interactive multi model(IMM)algorithm based on multi model.(2)In order to solve the problems that the existing tracking filtering and prediction algorithms need to match a separate filter for each model and make a priori assumption on the maneuvering state of the target before using the algorithm,a multi filter selection prediction(MFSP)algorithm is proposed.The algorithm solves the problem of filtering accuracy by identifying maneuvering targets and selecting appropriate filters according to different maneuvering intensities of targets.Simulation results show that,compared with the multi model algorithm,this algorithm has better filtering effect and more accurate prediction and estimation of target trajectory.(3)On the basis of the above research results,combined with the rich radar component library provided by System Vue,a complete three-dimensional air target tracking and prediction environment is built to test the multi axis model set matching and multi filter selection prediction algorithm proposed in this paper.The test results show that the model matching rate of the multi axis model set matching algorithm is higher in the complex maneuvering transformation of the target,and the multi filter selection prediction algorithm can filter and predict the target with different maneuvering intensity,which is suitable for the system with high accuracy of the predicted trajectory.