Research On Target Tracking Algorithm Based On Hybrid Filter Of UKF And PF

No matter in the military field or the civilian field,the target tracking technology plays an important role.It has important application value in the fields of enemy situation monitoring,sea,land and air defense,intelligent transportation,logistics tracking of cargo warehouses,etc.It has always been a research hotspot of domestic and foreign scholars.In order to improve the target tracking accuracy and improve the tracking effect,this paper presents a hybrid filtering target tracking algorithm based on the combination of unscented Kalman filter(UKF)and particle filter(PF).The main research contents are as follows:First,the paper expounds the research background and significance of this topic,and introduces the research status at home and abroad.Next,the existing linear Kalman filter(KF),extended Kalman filter(EKF),unscented Kalman filter(UKF),and particle filter(PF)algorithms are introduced in detail.On this basis,the advantages and disadvantages of the above algorithms are analyzed,and the causes of the errors of the four algorithms are introduced.In view of the problems and defects of the existing KF and PF algorithms,this paper proposes a target tracking algorithm based on UKF and PF hybrid filtering.First,use PF to initially estimate the state,it overcomes the constraints of the nonlinear system on the target tracking algorithm;then,in order to eliminate the effects of singular values and particle degradation on the tracking results,further improve the target tracking accuracy,UKF is applied to the estimation result of the previous step.On this basis,a single-observation target tracking model based on UKF algorithm,PF algorithm,UKF and PF hybrid filtering algorithm was established respectively.MATLAB simulation experiment results show that the single-observation target tracking algorithm based on UKF and PF hybrid filtering proposed in this paper has the highest tracking accuracy compared with the other two algorithms,which effectively improves the target tracking effect.In view of the shortcomings of measurement accuracy,range and reliability of a single observation station,this paper further studies and analyzes the application of filtering algorithms in multiple observation stations target tracking.First,the equal-weighted fusion method is used to establish the target tracking algorithm model of multiple observation stations based on UKF and PF.Then,an fitness function is established to improve the weighted fusion method,which is used to give different weights to each observation station.Finally,the fitness weighted target tracking algorithm of multi-observation station based on UKF and the fitness weighted target tracking algorithm model of multi-observation station based on UKF and PF hybrid filtering are established respectively.MATLAB simulation experiment results show that the fitness weighted fusion method proposed in this paper effectively improves the multi-observation target tracking fusion effect,and the multi-observation fitness weighted target tracking algorithm based on UKF and PF hybrid filtering has the best tracking effect among all the above algorithms and improves the accuracy of target tracking.Finally,on the basis of the above research,this article uses object-oriented development methods,uses the Qt development platform with excellent cross-platform characteristics under the Windows system,and combines Qt and MATLAB mixed programming technology to achieve target tracking simulation through C++ language programming system.The system simulation platform runs well under the premise of meeting the design requirements,realizes the intuitive display of the target tracking effect,and has certain reference value.