Research And Implementation Of Multi-target Tracking System Software Based On Passive Location

The target tracking system is usually able to track multiple targets in the detection area at the same time.It sets up the track of a new target from scratch and keeps the update of the track of multiple targets.It can adapt to the maneuvering behavior of the target to ensure a certain tracking accuracy.This thesis studies the multitarget tracking system.The system calculates the locating point based on the multi station TDOA passive location method.Then it tracks targets to which the location points belong.First,for the problem of target state estimation in target tracking system,this thesis studies the filtering model of Kalman filter algorithm.Through simulation experiments,it is found the filtering performance will decrease when tracking maneuvering targets.For the filtering problem of nonlinear systems,this thesis studies the nonlinear filtering algorithm including Extended Kalman filter,Unscented Kalman filter and Cubature Kalman filter.Research on the tracking performance of the three algorithms for the same nonlinear system target,the experiment shows Cubature Kalman filter algorithm has the highest tracking accuracy and filtering stability.Aiming at the tracking problem of maneuvering targets,combined with the advantages of the Interactive Multiple Model algorithm,this thesis presents the recursive process of Interacting Multiple Model algorithm based on Cubature Kalman filter.The simulation proves the algorithm is more suitable for maneuvering target tracking than single model filtering algorithm.Secondly,aiming at the track initiation problem in the dense multiecho environment,this thesis studies Intuitive method,Logical method and modified Hough transform method in track initiation algorithm.And it studies the initial performance of the three initial algorithms in a clutter environment.The experiment shows modified Hough transform method has a higher probability of track detection and a lower false alarm rate.On the single target track association problem in dense multiecho environment,this thesis studies the association principle and calculation steps of the Nearest Neighbor Data Association algorithm and the Probabilistic Data Association algorithm.The simulation shows the Probabilistic Data Association algorithm has lower root mean square error than the Nearest Neighbor Data Association algorithm when tracking a single target.Aiming at the multitarget track association problem,this thesis studies the Joint Probabilistic Data Association algorithm.Considering the faults of the Joint Probabilistic Data Association algorithm,it is improved based on multiple thresholds.To further reduce the number of candidate measurements in the tracking gate,thereby reducing the computational complexity of the algorithm,an ellipse tracking gate and a direction tracking gate are used to filter measurement data at the same time.The simulation shows the Improved Joint Probability Data Association algorithm based on Multiple Thresholds ensures the tracking accuracy.It correctly correlates the measurements of multiple cross moving targets.Besides,it avoids the phenomenon of track interleaving,and significantly reduces the algorithm execution time for high clutter density.Finally,based on the research,a multitarget tracking system software based on passive positioning is developed.The system realizes functions including data communication,locating point calculation,track initiation,track maintenance,track termination,situation drawing,area warning and historical track display.It also completes the positioning,tracking and situation visualization of flight targets in the detection area.It achieves the purpose of real-time monitoring and recording track.This system has completed all coding and testing work,and can cooperate with relevant base station equipment to carry out testing work in the field.