Study On Tracking And Data Association For Radar Detecting Space Target

Target tracking and data association are showing broad application prospects in both the military and civilian areas.And this thesis talks about the problem of tracking and data association for radar detecting space target,focusing on multi-target measure points rejection,track initiation,track maintenance,tracking filter and track termination,etc.In the multi-target radar detection process,because the target relative velocity and acceleration to radar are too large,this high d ynamic characteristic often causes wrong track association,resulting in unstable tracking or target lost.To solve this problem,multi-target data association methods for reducing space targets' dynamic characteristic influence are proposed.That is,based on space target d ynamic constraints,the difference between space target measurement data and prediction data by short-arc orbit determination is calculated to associate radar multi-target measurement data.In addition,there are two main processes in this thesis,with engineering practice taken into consideration,in the data association,Logic-Based Method is used to initiate a track,and Nearest Neighbour Algorithm is used to maintain it.In the tracking and filtering stage,the algorithm is usin g the sliding-window Least Square Method in order to fit the space target track and predict the location in the future space moments.The algorithm simulation has been completed using MATLAB.Moreover,in practical engineering applications,the algorithm also has been successfully applied to several radars and achieved good results by experiments.