Research On Nonlinear Target Tracking Fusion For Networked Radars

In recent years,nonlinear target tracking and data correlation technology have become the research hot spots of relevant experts and scholars,and have been widely used in military field or civil life.In practical applications,most of them are nonlinear systems.The choice of tracking filter algorithm is especially important,and it is the primary problem that needs to be solved in target tracking.The environment for modern radar survival is more and more complex.The data fusion of network radar can be used in the current situation.Comparing with single radar operation,the data of networked radar tracking fusion is more comprehensive and accurate,which is one of the development trends of radar detection technology in the future.This paper focuses on the fusion algorithm of networked radar tracking.The main works are as follows:1.For the distributed multi-sensor track fusion problem,in order to improve the track correlation rate,this paper improves the gray track association algorithm based on DS(Dempster-Shafer)theory.In the multi-sensor and multi-target tracking system,in order to reduce the influence of signal redundancy,the DS theory is introduced to fuse the observations at each moment,the correct recognition rate of the signal is improved,and the gray correlation degree between the sensors is calculated.Then,a valid double threshold is set with this correlation as the global statistic,and finally,the global optimal trajectory association decision is performed according to certain rules.The simulation results show that the performance and robustness of the proposed algorithm are better than the traditional gray track correlation algorithm in the scene with dense targets and many tracks,it has higher tracking accuracy.2.An improved algorithm based on unscented Kalman filter(UKF-LS)is proposed.Firstly,the algorithm principle and basic steps of extended Kalman filter(EKF)are studied,and the least squares estimation(LS)algorithm is introduced on this basis.In order to reduce the first-order truncation error of EKF algorithm in nonlinear systems,an improved algorithm based on unscented Kalman filter is proposed based on unscented transform.Finally,the tracking accuracy is calculated according to the state estimation value of the improved fusion algorithm.The simulation results show that the improved unscented Kalman filter(UKF-LS)algorithm proposed in this paper has better tracking performance,and the performance of EKF-LS algorithm is further improved.3.In the radar and infrared sensors tracking system,an improved radar and infrared sensors cooperative tracking algorithm is proposed,which effectively reduces the working time of the radar in the netted system.Firstly,the algorithm establishes a tracking model of radar and infrared sensor based on the interactive multi-model and unscented Kalman filter(IMMUKF)algorithm.Secondly,a new tracking quality factor is designed to control the radiation of the radar.The residual of the new information obtained by comparing the filtering result with the estimated measurement value is used as a standard to adaptively control the working time of the radar and the infrared sensor to switch of radar and infrared sensors.The simulation results show that the algorithm can reduce the radar radiation time with good tracking accuracy.