对地观测卫星成像调度的多目标约束修正方法  

A Multi-objective Constraint Modification Approach for Imaging Scheduling of Earth Observing Satellite

作　　者：于海[1] 李军[1] 王钧[1] 景宁[1] 

机构地区：[1]国防科学技术大学电子科学与工程学院,湖南长沙410073

出　　处：《航空学报》2009年第3期512-517,共6页Acta Aeronautica et Astronautica Sinica

基　　金：国家"863"计划(2007AA12Z229;2007AA120200);国家自然科学基金(60604035)

摘　　要：约束修正是对地观测卫星成像调度的重要组成部分,负责处理成像调度方案的约束满足与优化,是一类复杂的组合优化问题。为得到优化可行的成像调度方案,提出一种新的约束修正方法。针对约束修正问题构建时间序有向图模型,并将约束修正问题归结为点带约束成本的优化路径搜索问题;在此基础上,提出一种基于标记更新的多目标约束修正算法。实际问题的实验与分析表明:该方法能够在规定的时间内求得问题所有的多目标优化解,有效地解决了卫星成像调度的约束满足与优化问题。Constraint modification is an important part of imaging scheduling for earth observing satellite. It deals with constraint satisfaction and optimization of the imaging scheduling, and is a complicated problem of optimized combination. This article proposes a new constraint modification approach to acquire the optimal and workable imaging scheduling. Aiming at the constraint modification problem, a time ordered directed graph model is constructed. The constraint modification is also summed up as an optimal path searching problem of node with cost. Furthermore, a multi-objective constraint modification algorithm based on label-setting is proposed. Analysis of experiments with practical problems indicates that the proposed approach is capable of finding all Pareto optimal solutions and effectively resolves the constraint modification problem.

关 键 词：卫星 成像调度 多目标 约束修正 时间序有向图 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]