Research On Group Target Tracking And Shooting Assistant Decision Technology

Cluster attack system has high intelligence,quantitative advantages and powerful power,which poses a great threat to air defense system,and has attracted great attention of defense system researchers at home and abroad.In the modern war scene,the distribution of group target measurement data obtained by radar may be very dense,so it is difficult to distinguish each target,and the multi-target data association and tracking process will become extremely difficult.In the process of penetration,enemy targets often have structural changes such as group separation or combination,which will lead to the instability of group target tracking accuracy.Therefore,for dense and indistinguishable group targets,it is of great military significance to adopt group target tracking algorithm and group overall structure state detection algorithm,and work together with related weapon systems.Based on the above research background,this paper studies and analyzes group target tracking and shooting assistant decision-making.The main research contents are as follows:Firstly,the group target measurement division method and track initiation algorithm are studied.The DBSCAN clustering algorithm is used to divide the radar measurement data,which solves the problem that the number of group targets cannotbe determinedin the actual combat environment.Aiming at the problem of group target track initiation,a group target track initiation algorithm based on nearest neighbor and graph theory is proposed.The connected graph is constructed according to the measurement distribution characteristics of group center track at the initial continuous time and the nearest neighbor principle.The weighted adjacency matrix is established through the graph information,and the initial track of group target is determined and optimized by depth first traversal and screening rules,It not only improves the correct track initiation rate and reduces the false track initiation rate,but also ensures the overall accuracy of the group target initiation track.Secondly,the group target tracking algorithm and the group target separation and combination detection algorithm are studied.Using group targetmodel,group target tracking gate and correlation gate,combined with joint probabilistic data association algorithm and Kalman filter algorithm,single group target tracking filter and multi group target tracking filter are designed.Aiming at the problem of detecting the change of the overall structure of the group,a group target separation and combination detection algorithm based onDBSCAN clustering is proposed.The algorithm judges the possibility of group separation and merging according to the group target model,the number of measurements falling into the group target tracking gate and the number of predicted group centers in the intersection area of the group target association gate;The DBSCAN algorithm is used to cluster the measurements in the group target association gate to further determine whether the group targets are separated or merged;According to the equivalent measurement variance in the new group target tracking gate,the corresponding group target tracking gate size and the number of group members are calculated and updated to ensure the real-time performance of group target separation and detectionand improve the tracking accuracy before and after separation or combination.Thirdly,the auxiliary decision-making scheme of air group target combat shooting is designed.Aiming at the problem of multi group target threat evaluation,the appropriate evaluation indexes are selected,each index is quantitatively described by fuzzy theory,the weight of each quantitative factor is determined by analytic hierarchy process,and the threat degree is calculated by weighting each threat index of group target.In order to distribute the fire power to the penetration group targets,the damage probability of the fire unit to the group targets is calculated by using the damage probability model of the future airspace window,and the threat degree and damage probability of the group targets are combined to construct the objective function,which is optimized by the artificial fish swarm algorithm,which can effectively determine the priority strike target of each fire unit,Provide the best shooting assistant decision-making scheme.Finally,a hardware in the loop simulation platform for group target tracking and shooting assistant decision-making is built based on Windows system and QT framework.The three-dimensional route simulation module,radar simulation module,measurement module,target tracking module,shooting auxiliary decision-making module and fire control solution module are designed respectively.Throughthe joint debugging of the system,the group target tracking algorithm and shooting assistant decision-making are tested,the expected effect is achieved,and the real-time and feasibility of the system are verified.