成像卫星密集任务合成方法及其调度算法  

Intensive task merging method and scheduling algorithm for imaging satellites

作　　者：于静 杨文沅 刘晓路 邢立宁[1] YU Jing;YANG Wenyuan;LIU Xiaolu;XING Lining(College of Systems Engineering,National University of Defense Technology,Changsha 410073,China)

机构地区：[1]国防科技大学系统工程学院,湖南长沙410073

出　　处：《华中科技大学学报（自然科学版）》2021年第10期73-78,共6页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金资助项目(61773120);国家优秀博士论文基金资助项目(2014-92);湖南省智能物流技术重点实验室基金资助项目(2019TP1015)。

摘　　要：当一般卫星面临问题规模较大且任务比较密集时,传统的调度模型会出现任务排斥,造成观测效率及观测收益都较低的现象.针对该问题,提出了基于任务合成机制的多星调度算法.首先,考虑任务之间的约束条件,建立基于均值漂移的卫星任务合成算法;然后,考虑卫星资源的固存约束、能量约束,以及观测任务之间的观测时间、观测角度等约束条件,建立了基于均值漂移的多星任务合成调度问题模型;最后,结合任务合成算法及问题特点,用改进的蚁群求解算法进行求解,并设计了Insert搜索算子来提高算法的探索能力.仿真实验验证了该任务合成方法及求解算法的效率.The traditional scheduling model often occurs mutual exclusion between tasks,resulting in low observation efficiency and observation benefits,when general satellites face larger problems and more intensive tasks.To solve the above problem,a multi-satellite scheduling algorithm was presented based on the task merging mechanism.First,the satellite merged task algorithm was established based on meanshift,considering the constraints between tasks.Then,the multi-satellite merged task scheduling problem model was established,taking into account the constraints on the storage and the energy of satellite resources,and the observation time,the observation angle and other constraints between tasks.Finally,by combining the merged task algorithm and the characteristics of this problem,an improved ant colony algorithm was used to solve this scheduling problem,and the Insert operator was designed to improve the exploration of the algorithm.Simulation results verified the efficiency of the merged task method and the proposed algorithm.

关 键 词：成像卫星 任务合成 任务规划 均值漂移 蚁群算法 密集任务 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]