Research On The Optimal Maneuvering Trajectory Of UAV Cluster Target Localization And Tracking

At present,with the continuous deepening of UAV technology research and the continuous improvement of industrial manufacturing level,UAVs have broad application prospects in both military and civilian aspects.In the field of target tracking,UAVs have many applications in search and rescue,reconnaissance,surveillance,and tracking strikes due to their easy deployment and low cost.In military scenarios,with the complexity and informatization of the modern battlefield environment,the combat capability of a single UAV is increasingly insufficient,while the UAV cluster has the characteristics of large-scale and multi-function,which can lead in a systematic way of combat The orientation of future wars.Therefore,the related research on target positioning and tracking of UAV clusters has extremely important practical value.It is not only an important prerequisite for the follow-up of UAV clusters on target hunting and attacking and chasing and escaping confrontation problems,but also an important research on "swarm intelligence".The foundation is the core research direction of future autonomous intelligent unmanned systems.This thesis mainly studies the optimal maneuver trajectory of UAV cluster target locating and tracking.The main work done is as follows:(1)Established the mathematical model of the multi-UAV target tracking system,including several typical target motion models and multi-sensor measurement models based on bearing-only and range-only.Secondly,the Unscented Kalman Filter(UKF)algorithm and process for UAV cluster target tracking are given.Finally,the related performance evaluation indexes of Monte Carlo simulation method and filtering algorithm are given: Cramér–Rao lower bound(CRLB),root mean square error(RMSE),etc.(2)Aiming at the problem of UAV cluster target locating,a GDOP index that can be used to evaluate the influence of the geometric configuration between the UAV measurement sensor and the target on the locating accuracy is studied,and based on bearing-only locating and range-only locating Under GDOP,the form of the optimal geometric configuration that the UAV sensor and target can reach and its necessary and sufficient conditions are given.(3)Aiming at the multi-target tracking problem of UAV clusters,a non-linear integer programming mathematical model of UAV target allocation based on minimizing the sum of squares of GDOP differences is constructed,and corresponding solutions are given.Next,an optimal deployment plan for UAVs based on minimizing the sum of moving distances of UAVs is proposed,and this deployment scheme is summarized as a nonlinear 0-1 integer programming problem,and a penalty function-based Particle swarm algorithm is used to solve this UAV mission planning problem.(4)Aiming at the problem of the optimal maneuvering trajectory of the UAV cluster target tracking,a trajectory planning scheme based on minimizing the sum of the moving distance of the UAV is proposed,and the single-step optimal maneuvering strategy of the UAV is determined.Based on the UAV target allocation results and the optimal deployment plan,the specific algorithm flow of the UAV cluster single target and multi-target locating and tracking is given.The simulation analysis of the target cluster tracking under the two UAV target locating modes,bearing-only and range-only,verifies the effectiveness of the proposed method.