Cooperative Mission Planning for UAV Based on Improved Wolf Pack Algorithm
The use of multiple Unmanned Aerial Vehicles(UAV)in modern warfare is increasing,making UAV mission planning crucial for intelligent UAV operations.In response to the UAV task allocation problem under subsystem capability constraints,this study proposes a chaos reverse learning wolf pack optimization algorithm,CRL-AMIWPA,based on Levy Flight(LF)optimization and auction mechanism.First,the UAV capability matrix and task scenario are defined.The heterogeneity of the UAVs,task execution capability,and minimum task execution requirements are described in the same matrix.The objective function is established by considering the weighted sum of distance fuel consumption and latest task completion time,and the task allocation model is established under subsystem constraints.Subsequently,the individual coding of the wolf pack is designed,whereby each coding scheme represents a task allocation strategy.A corrective auction strategy based on contract network is used for solutions that fail to satisfy the minimum task capability requirements.In addition,during the initialization stage of the wolf pack,the Tent chaos operator and reverse learning strategy are applied to distribute the wolf pack individuals evenly in the solution space and improve the diversity of the initial population.Finally,LF strategy is used to optimize the search process,thereby enhancing the ability to escape local optima.Simulation results demonstrate that the proposed algorithm effectively solves the UAV task allocation problem under subsystem-constraint scenarios and exhibits better optimization performance and convergence speed than other swarm intelligence and Wolf Pack Algorithms(WPA).
Wolf Pack Algorithm(WPA)task allocationchaotic optimizationvariable step size optimizationswarm intelligence algorithm
国家科技期刊平台
NSTL
万方数据
