| 引用本文: | 刘立昊,董正宏,苏昊翔,陈公政.博弈协商策略下的多星分布式协同任务规划[J].控制理论与应用,2023,40(3):502~508.[点击复制] |
| Liu Li-hao,Dong Zheng-hong,Su Hao-xiang,Chen Gong-zheng.Multi-satellite distributed mission scheduling via game strategy[J].Control Theory and Technology,2023,40(3):502~508.[点击复制] |
|
| 博弈协商策略下的多星分布式协同任务规划 |
| Multi-satellite distributed mission scheduling via game strategy |
| 摘要点击 2542 全文点击 645 投稿时间：2021-04-23 修订日期：2022-07-10 |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| DOI编号 10.7641/CTA.2021.10340 |
| 2023,40(3):502-508 |
| 中文关键词 卫星 分布式系统 任务规划 博弈 优化算法 |
| 英文关键词 satellite distributed system mission scheduling game theory optimization algorithm |
| 基金项目 国家自然科学基金项目(61602516)资助. |
|
| 中文摘要 |
| 随着星载智能水平、航天运载能力、星间通信能力的全面提升, 智能化、组网化、自治化的卫星系统成为当
前发展趋势, 传统的集中管控方式已难以适应未来分布式卫星系统的综合管理需求. 针对分布式遥感卫星系统任务
规划问题, 本文提出了一种博弈协商机制的多星自主分布式任务规划模型. 在该模型中, 每颗卫星作为“理性”个体
参与任务规划, 在每轮博弈中基于局部目标信息和全局交互信息, 利用自适应粒子群优化算法不断更新自身的“行
动”, 直至达到系统平衡. 仿真结果显示, 分布式任务规划方法能够灵活地应对不同规模的问题场景, 算法性能不会
随着问题规模的增大而出现显著的下降, 有效地克服了传统方法在大规模任务规划场景中优化时间激增、收敛速
度慢等缺点, 能高效地获取全局性能较优的规划解. |
| 英文摘要 |
| With the advancement of on-board intelligence, space carrying capacity, and inter-satellite communication
capabilities, intelligent, networked and autonomous satellite systems have become the current development trend, and traditional
centralized management and control methods are difficult to adapt to the integrated management needs of future
distributed satellite systems. Aiming at the missions scheduling of distributed remote sensing satellite systems, this paper
proposes a distributed mission planning model for remote sensing satellites based on the game theory. In this model, each
satellite participates in mission planning as a “rational” player. In each round of the game, based on the local target information
and the global interactive information, the adaptive particle swarm optimization algorithm is used to continuously
update player’s own “action” until the system balance is reached. The simulation results show that the distributed missions
scheduling method can flexibly deal with problem scenarios of different scales, and the performance of the algorithm will
not decrease significantly with the increase of the problem scale, and can efficiently obtain the global optimal solution. |
|
|
|
|
|