改进PSO算法在多无人机协同任务分配中的应用

针对多无人机协同任务分配越来越复杂的问题，采用一种改进的阶层分级粒子群优化算法（HGIWPSO）获得最优分配方案。首先，根据粒子适应度值将种群动态划分为三个不同阶层，依据不同阶层粒子特性选择合适的学习模型，并引入独立权重思想调节惯性权重大小，平衡算法全局与局部搜索能力，提高算法性能；然后，建立协同多任务分配问题模型，采用多余负载竞拍方案减少非法劣解，通过实数编码建立粒子和实际分配方案之间的映射关系，解决实际分配问题。实验结果表明，该算法能够有效解决复杂约束条件下多无人机协同任务分配问题，得到最优分配序列，具有一定的理论以及实际意义。

Application of improved PSO algorithm in multi UAV cooperative task allocation

In order to solve the increasingly complex problem of multi-UAV collaborative task allocation, this paper used an improved hierarchical particle swarm optimization algorithm(HGIWPSO) to obtain the optimal allocation scheme. Firstly, according to particle fitness, it divided the population dynamics into three different classes, and selected the appropriate learning model according to the characteristics of different classes of particles, and introduced the independent weight idea to adjust the weight of the inertia weight that improved the performance of the algorithm. Then, it established the model of collaborative multi task assignment problem, and adopted the redundant load auction scheme to reduce illegal solutions. It established the mapping relationship between the particle and the actual allocation scheme through real number coding to solve the actual allocation problem. Finally, the experimental results show that the algorithm can effectively solve the problem of multi-UAV coordinated task assignment under complex constraints and obtain the optimal distribution sequence. It has certain theoretical and practical significance.