摘要
多无人机执行广域搜索攻击任务下的协同多任务优化分配需要满足多类复杂约束。在通用无人机协同多任务分配模型的基础上,综合考虑了包括任务时间约束和无人机弹药消耗等多类复杂约束条件,提出了一种基于离散粒子群算法的多无人机协同多任务分配方法。根据协同多任务分配问题的特点,将多任务分配的任务时序约束和多机协同约束融入到算法的粒子矩阵编码中,将无人机弹药约束和任务时间约束融入到粒子更新的过程中,设计了符合实际问题离散域特点的粒子位置和速度更新的交叉策略和变异策略。仿真结果表明,上述算法能在满足多类复杂约束的条件下有效地解决无人机作战目标协同多任务优化分配问题。
To solve the problem of cooperative multi -task assignment （CMTAP） for unmanned aerial vehicle （UAV） under wide area search and attack mission, a discrete particle swarm optimization algorithm （DPSO） is pro- posed in the paper. The algorithm not only considers task precedence, coordination and time constraints, but also holds the view of the dynamic task execution time window constraints and the limited weapons of each UAV that de- plete with use. According to the characteristics and constraints of CMATP, particles were coded as a matrix to satisfy the constraints of task precedence and coordination. The cross and mutation strategies for the position and speed up- dating of particles were applied to make the DPSO algorithm more suitable for solving this discrete area problem. Strategies of self - organization inertia weight and acceleration coefficient were introduced to make full use of the glob- al search capability and effectively overcome the PSO disadvantages of slow convergence and easy to fall into local op- timum. Simulation results demonstrate the feasibility and efficiency of the proposed DPSO algorithm.
出处
《计算机仿真》
北大核心
2018年第2期22-28,共7页
Computer Simulation
关键词
无人飞行器
多任务分配
组合优化
离散粒子群
广域搜索攻击
Unmanned aerial vehicle（UAV）
Multi -task assignment
Combinatorial optimization
Discrete particle swarm
Wide area search and attack
