Research On Data Association Technology In Target Tracking

Target tracking technology is widely used in military and civilian fields,and data correlation technology in target tracking technology can achieve the correlation between target points and tracks to achieve accurate tracking of targets.With the rapid development of modern aerospace technology,the accuracy requirements of data association in target tracking are also getting higher and higher,especially in the complex environment,multi-target tracking is prone to leakage and mismatch,so the research on multi-objective data association algorithm has important practical significance in achieving accurate target tracking.Based on the analysis of existing data association algorithm,this thesis focuses on the multi-objective data association algorithm.Aiming at the shortcomings of existing algorithms,a Rough Set Probabilistic Data Association Algorithm for single radar and a based on Interactive Multiple Model Optimization Multiple Hypothesis Tracking Algorithm for multi-radar are proposed to improve data association accuracy,the results achieved in this paper mainly include the following aspects:First of all,in view of the problem that the existing single radar multi-target tracking Probability Data Association Algorithm does not consider the processing of two or more targets tracking gate intersection regions and the Joint Probability Data Association Algorithm constructing joint events can become quite complicated and may lead to combined explosions,so an based on the Rough Set Probability Data Association Algorithm is proposed.The algorithm considers the influence of public measurement on data association.By classifying the measurement using rough set theory,the accuracy of data association and the real-time tracking of target tracking in dense clutter environment are guaranteed.Secondly,the existing multi-radar data association algorithm uncertainties target motion model and Multiple Hypothesis Tracking Algorithm produces a problem of exponential growth when the target and clutter increase,a based on Interactive Multiple Model Optimization Multiple Hypothesis Tracking Algorithm is proposed,by introducing an interactive multi-model filter to estimate the target motion model,the error caused by target maneuvering is reduced,and the radial velocity information is added to correct the path confidence parameter when the hypothesis is generated,which reduces the generation of the hypothesis and improves the correlation accuracy of multi-radar joint data association.Finally,the target tracking system based on radar is modeled by using SystemVue platform,and the above research results were simulated and tested.The simulation results show that compared with the existing algorithms,the proposed algorithm can significantly improve the accuracy of data association in the target tracking system model.