复杂场景下考虑能源特性的多太阳能无人机任务分配方法  

Multi solar-powered UAV task assignment method considering energy characteristics in complex scenarios

作　　者：张尚义 李文光[1] 葛佳昊 Zhang Shangyi;Li Wenguang;Ge Jiahao(School of Aerospace Engineering,Beijing Institute of Technology,Beijing 100081,China;School of Aeronautic Science and Engineering,Beihang University,Beijing 100191,China)

机构地区：[1]北京理工大学宇航学院,北京100081 [2]北京航空航天大学航空科学与工程学院,北京100191

出　　处：《战术导弹技术》2024年第5期99-110,共12页Tactical Missile Technology

摘　　要：针对太阳能无人机任务分配与能源特性耦合、建模与算法定制化问题,基于太阳能无人机能源特性和复杂应用场景进行问题建模,建立了新的适应度指标及约束取代传统指标和约束。提出了分时工作模式,在无人机目标分配、目标执行顺序之外增加了起飞时间这一新的待分配要素。重新设计了定制化双染色体编解码模式以兼顾粒子对起飞时间的表达。基于遗传学习对粒子群算法的最优粒子生成策略进行改进,给出了改进定制粒子群算法的求解步骤。结果表明,所提改进定制粒子群算法在求解太阳能无人机任务分配问题上是有效可行的。Aiming at the problem of coupling the task assignment of solar-powered UAVs with energy characteristics and the need for customization of modeling and algorithms,the problem is modeled based on the energy characteristics of solar UAVs and the complex application scenarios,and new adaptability indexes and constraints are established to replace the traditional indexes and constraints.A time-sharing work model is proposed,and a new element is assigned for the takeoff time in addition to the target assignment and target execution sequence of the UAVs.A customized dual-chromosome coding and decoding model is redesigned to take into account the particle representation of takeoff time.The optimal particle generation strategy in the particle swarm algorithm is improved based on genetic learning,and the solution steps of the improved customized particle swarm algorithm are given.The simulation results show that the proposed improved customized particle swarm algorithm is effective and feasible in solving the solar UAV task assignment problem.

关 键 词：太阳能无人机 任务分配 能源特性 复杂场景 多目标优化 遗传学习 粒子群算法 

分 类 号：V249[航空宇航科学与技术—飞行器设计]