多星协同观测遗传-演进双层任务规划算法  

Genetic-evolutionary bi-level mission planning algorithm for multi-satellite cooperative observation

作　　者：李阳阳 罗俊仁 张万鹏 项凤涛 LI Yangyang;LUO Junren;ZHANG Wanpeng;XIANG Fengtao(College of Intelligence Science and Technology,National University of Defense Technology,Changsha 410073,China)

机构地区：[1]国防科技大学智能科学学院,湖南长沙410073

出　　处：《系统工程与电子技术》2024年第6期2044-2053,共10页Systems Engineering and Electronics

基　　金：国家自然科学基金(U1734208);湖南省自然科学基金(2021JJ40693)资助课题。

摘　　要：多星协同任务规划方法是天基卫星系统管控的关键支撑。围绕多星协同对地观测任务展开分析,首先建立多星协同任务规划模型,包括卫星轨道参数、约束条件和待观测目标点等;其次设计了遗传演进双层求解架构,将多星任务规划问题拆解为顶层多星任务分配问题和底层单星任务调度问题,上层采用基于引导的多种群遗传算法(multi-population genetic algorithm,MPGA),将启发式结果融入到任务分配算法中,下层采用改进遗传算法对单星任务调度问题进行求解;最后针对适用性问题,设定随机和均匀分布两组目标,采用不同卫星数量设计实验验证了遗传演进双层求解框架的有效性。Multi-satellite cooperative mission planning method is a key node in the space-based satellite system architecture.Firstly,the multi-satellite cooperative for Earth observation mission is analyzed and a multi-satellite cooperative mission planning model is established,including satellite orbit parameters,constraint conditions,and target points to be observed.Then,a genetic-evolutionary bi-level solution architecture is designed,which decomposes the multi-satellite mission planning problem into a top multi-satellite mission assignment problem and a bottom single-satellite mission scheduling problem.The upper level uses the guided multi-population genetic algorithm(MPGA)to integrate the heuristic results into the task allocation algorithm,and the lower level uses the improved genetic algorithm to solve the single-satellite task scheduling problem.Finally,aiming at the applicability problem,two groups of objectives are set randomly and uniformly distributed,and experiments are designed with different numbers of satellites to prove the effectiveness of the genetic-evolutionary bi-level solution framework.

关 键 词：卫星任务规划 遗传演进架构 多种群遗传算法 并行算法 

分 类 号：V19[航空宇航科学与技术—人机与环境工程]