Management and mission planning over a swarm of unmanned aerial vehicle (UAV) remains to date as a challenging research trend in what regards to this particular type of aircrafts. These vehicles are controlled by a number of ground control station (GCS), from which they are commanded to cooperatively perform different tasks in specific geographic areas of interest. Mathematically the problem of coordinating and assigning tasks to a swarm of UAV can be modeled as a constraint satisfaction problem, whose complexity and multiple conflicting criteria has hitherto motivated the adoption of multi-objective solvers such as multi-objective evolutionary algorithm (MOEA). The encoding approach consists of different alleles representing the decision variables, whereas the fitness function checks that all constraints are fulfilled, minimizing the optimization criteria of the problem. In problems of high complexity involving several tasks, UAV and GCS, where the space of search is huge compared to the space of valid solutions, the convergence rate of the algorithm increases significantly. To overcome this issue, this work proposes a weighted random generator for the creation and mutation of new individuals. The main objective of this work is to reduce the convergence rate of the MOEA solver for multi-UAV mission planning using weighted random strategies that focus the search on potentially better regions of the solution space. Extensive experimental results over a diverse range of scenarios evince the benefits of the proposed approach, which notably improves this convergence rate with respect to a na\"ive MOEA approach.

翻译：无人机群的管理与任务规划至今仍是该特定类型飞行器领域一项具有挑战性的研究趋势。这些飞行器由多个地面控制站（GCS）操控，被指令在特定地理利益区域内协同执行不同任务。从数学角度看，协调并分配任务给无人机群可建模为约束满足问题，其复杂性和多个相互冲突的准则促使学界采用多目标求解器，例如多目标进化算法（MOEA）。编码方法由代表决策变量的不同等位基因构成，而适应度函数则负责检查所有约束是否满足，同时最小化问题的优化准则。在涉及多任务、多无人机和多地面控制站的高复杂度问题中，由于搜索空间相较于有效解空间极为庞大，算法的收敛速度显著增加。为解决此问题，本文提出一种加权随机生成器，用于创建和变异新个体。本文的主要目标是利用加权随机策略，将搜索聚焦于解空间中潜在更优的区域，从而降低面向多无人机任务规划的多目标进化算法（MOEA）求解器的收敛速度。针对多样化场景的大量实验结果证明了所提方法的优势，其相较于朴素多目标进化算法（MOEA）方法显著提升了收敛速度。