系统工程与电子技术 ›› 2022, Vol. 44 ›› Issue (5): 1600-1608.doi: 10.12305/j.issn.1001-506X.2022.05.21

• 系统工程 • 上一篇    下一篇

基于分布式协同进化的星座自主任务规划算法

高天旸1,2,*, 胡笑旋1,2, 夏维1,2   

  1. 1. 合肥工业大学管理学院, 安徽 合肥 230009
    2. 过程优化与智能决策教育部重点实验室, 安徽 合肥 230009
  • 收稿日期:2021-09-22 出版日期:2022-05-01 发布日期:2022-05-16
  • 通讯作者: 高天旸
  • 作者简介:高天旸(1996—), 男, 硕士研究生, 主要研究方向为卫星任务规划|胡笑旋(1978—), 男, 教授, 博士, 主要研究方向为空间信息网络任务规划与资源调度|夏维(1983—), 男, 副教授, 博士, 主要研究方向为智能决策
  • 基金资助:
    国家自然科学基金(72071064)

Constellation autonomous mission planning algorithm based on distributed co-evolution

Tianyang GAO1,2,*, Xiaoxuan HU1,2, Wei XIA1,2   

  1. 1. School of Management, Hefei University of Technology, Hefei 230009, China
    2. Key Laboratory of Process Optimization and Intelligent Decision-making, Ministry of Education, Hefei 230009, China
  • Received:2021-09-22 Online:2022-05-01 Published:2022-05-16
  • Contact: Tianyang GAO

摘要:

星座协同自主任务规划是卫星自主化管理与控制技术中的重要一环。首先, 提出了一种分布式星座协同迭代优化策略, 星座内各星作为独立智能体通过“接收”“更新”“发布”的三阶段协作行为共同参与对整体任务方案的协调寻优。其次, 在该策略的基础上设计了一种分布式协同进化算法, 通过分布于不同卫星的多个亚种群在信息交互中并行进化以持续优化各星方案组合。最后, 在S698PM嵌入式开发环境下进行仿真实验, 通过与贪婪算法、集中式遗传算法以及CPLEX的对比测试, 验证了所提方法在恶劣通信环境下与大规模问题中的适用性和有效性。

关键词: 遥感星座, 自主规划, 分布式, 协同进化

Abstract:

Constellation collaborative autonomous mission planning is an important part of satellite autonomous management and control technology. Firstly, this paper proposes a distributed constellation collaborative iterative optimization strategy, in which each satellite in the constellation participates in the coordinated optimization of the system mission plan as an independent agent through the three-phase cooperative behavior of "receive", "update", and "release". Secondly, a distributed co-evolution algorithm is designed on the basis of this strategy, which allows multiple sub-populations distributed on different satellites to parallelly evolve with information interactions and continuously optimize the combination of plans for each satellite. Finally, this paper conducts simulation experiments in the S698PM embedded development environment. Through comparative tests with the greedy algorithm, the centralized genetic algorithm and CPLEX, the applicability and effectiveness of the proposed method in harsh communication environments and large-scale problems are verified.

Key words: remote sensing constellation, autonomous planning, distributed, co-evolution

中图分类号: