多基地多无人机协同侦察问题研究

 充分考虑侦察目标的侦察分辨率要求和侦察时间窗约束，以及位于不同基地的无人机(UAV)的侦察性能和可用数目，首次建立了更加贴近军事应用实际的多基地多UAV协同侦察问题(M-MUCRP)的数学模型，并提出了解决该模型的多基地多UAV协同侦察进化算法(M-MUCREA)。M-MUCREA的染色体数据结构有效地表达了问题的解，有利于交叉、变异等进化操作；充分利用与目标侦察分辨率要求以及目标位置和时间窗约束相关的启发信息，构造初始种群,避免进化过程收敛太慢；基于Pareto最优概念的选择算子确保解在多个目标上的有效优化；精英策略避免了丢失进化过程中产生的非劣解，加快算法收敛；变异和交叉算子在保证有效解的前提下，实现了解的多样性，避免了算法陷入局部最优。仿真实验验证了算法能够有效解决M-MUCRP。

Research on Multi-base Multi-UAV Cooperative Reconnaissance Problem

A multi-objective optimization model multi-base multiple UAVs cooperative reonnaissance problem(M-MUCRP)is presented for multi-base multiple unmanned aerial vehicles(UAVs)cooperative reconnaissance problem,which is more suitable for military applications.M-MUCRP takes the reconnaissance resolution demands and reconnaissance time window constraints into account,as well as the limited number of UAVs with different capabilities located in multiple bases.Then a multi-base multi-UAV cooperaitve recomaissance evolutionary algorithm(M-MUCREA)is proposed.M-MUCREA introduces effective chromosome representation which facilitates crossover and mutation operations.The heuristic information about the targets reconnaissance demands and locations are fully explored to construct initial feasible solutions and prevent the algorithm from converging too slowly.Pareto optimality based selection ensures the efficient multi-objective optimization effort of the algorithm.Elitism mechanism is adopted to prevent losing non-dominated individuals generated during the evolutionary process and speed up the convergence of the algorithm.Problem specific crossover and mutation operators ensure the feasibility of the children and diversify the population so as to prevent the algorithm from falling into local optimum.Simulation results show that the proposed algorithm can solve the problem effectively.