Research On Task Assignment Technology Of UAV Swarm Reconnaissance System

With the characteristics of small size,non-metallic material and low speed flight,micro-UAV can effectively reduce the probability of being detected by radar,and has become a new radar reconnaissance tool.The UAV swarm overcomes the insufficiency of the performance and load capacity of a single UAV,and by virtue of its far superior reconnaissance capability,it can efficiently complete the complex radar reconnaissance missions.By coordinating the matching relationship between the UAVs and the tasks,the rational assignment of resources can be realized.In this paper,the task assignment optimization in the mode transfer of phased array radar in UAV swarm reconnaissance is studied.The model and method of task assignment in UAV swarm are studied theoretically and verified by simulation.The main research contents are as follows:(1)According to the coverage mode of phased array radar beam position,three selection schemes of beam position reconnaissance point are analyzed and compared.On this basis,considering the limited on-board energy of the UAV,in order to save flight energy consumption,and shorten the duration of the mission,an optimal reconnaissance point selection algorithm is proposed,so that the UAVs can complete the whole beam positions reconnaissance mission by visiting the least reconnaissance points.According to the constraints of reconnaissance point access,UAV range and energy consumption,and aiming to minimizing the total energy consumption of UAVs,a multiple traveling salesmen optimization model for UAV swarm reconnaissance task assignment is established.(2)An improved partheno genetic algorithm-particle swarm optimization(PGA-PSO)is proposed to solve the task assignment problem under the goal of minimum energy consumption.The group optimal individual selection and multiple mutation operation are used to ensure the fast convergence of the algorithm and improve the solving precision of the algorithm.The simulation results verify the feasibility of the proposed algorithm and the efficiency of the selection of reconnaissance points,and the comparison among various algorithms shows that the improved PGA-PSO algorithm has both good global search capability and fast convergence speed.(3)In order to meet the demand of rapid operation in battlefield environment,the problem of cooperative assembly task assignment in UAV swarm reconnaissance is studied.Firstly,the beam position reconnaissance sequence of each UAV is obtained by the improved PGA-PSO algorithm under the optimization goal of shortest UAV reconnaissance time.Then k-degree smoothing algorithm is used to obtain the length interval of each UAV path after smoothing.Finally,the final standard path length of each UAV is determined according to the cooperative assembly strategy.The simulation verifies the feasibility of multi-UAVs completing the reconnaissance mission with the shortest time by using the cooperative assembly strategy under three beam position reconnaissance point selection schemes.