Research On Target Tracking And Sensor Information Fusion Based On Airborne Radar

Unmanned Aerial Vehicle(UAV)technology has gradually attracted extensive attention in military and civil fields because of its advantages such as long-distance combat and jamming,long-time reconnaissance and cruise,low-cost manufacturing and maintenance.As the core part of UAV sensor system,airborne radar is the top priority to be considered when deploying almost all UAV mission indicators.Taking target tracking based on airborne radar measurement information as the research background,starting from three aspects: complex electromagnetic occlusion,nonlinear moving target tracking performance and distributed sensor information fusion,this paper makes corresponding improvements for single target Bayesian filter and generalized label multi Bernoulli filter to adapt to various complex application scenarios,The main research contents are as follows:1.Aiming at solving the key scientific problems such as measurement loss and track interruption caused by Doppler Blind Zone of airborne early warning radar,a multi-mode airspace target tracking method based on Doppler Blind Zone is proposed.This method combines the root mean square embedded Cubature Kalman filter algorithm and particle filter algorithm,and sets four different working modes according to the tracking target state,in order to enhance the continuity and accuracy of track estimation in the blind zone.At the same time,according to the characteristics of high maneuverability of airspace targets,the interactive multi model algorithm is added to the filtering algorithm.The effectiveness of the proposed method is verified by the analysis of experimental results such as running time and minimum mean square error estimation.2.Aiming at the problem of nonlinear moving multi-target tracking,the δ-generalized label multi Bernoulli filter algorithm based on random set theory is studied,and the root mean square embedded Cubature Kalman filter algorithm is added to its Gaussian mixture implementation to improve it.A comparative experiment with the traditional algorithm is carried out under the nonlinear motion model.The OSPA value and running time value of each algorithm are compared and analyzed to verify the effectiveness of the proposed method.3.Aiming at the multi-target tracking problem in multiple airborne radar sensor working scenes,firstly,the implementation scheme of the traditional covariance crossover algorithm for single target tracking is introduced.Combined with the factors to be considered in the multitarget tracking scene,the defects and hidden dangers of the traditional covariance crossover method directly applied to the multi-target tracking filter are explained.An improved distributed radar sensor multi-target tracking method based on spatio-temporal two weight pre division is proposed.Through comparative experiments with single sensor and traditional multi-sensor information fusion scenarios,OSPA analysis is carried out under different clutter density.Finally,the effectiveness of the proposed method is verified.