宇航学报

宇航学报 ›› 2015, Vol. 36 ›› Issue (5): 583-588.doi: 10.3873/j.issn.1000-1328.2015.05.013

• 电子信息 • 上一篇    下一篇

地球观测网络成像任务可调度性预测方法

刘嵩，白国庆，陈英武   

  1. 1. 国防科学技术大学信息系统与管理学院，长沙410073；2.驻沈阳铁路局长春军事代表办事处，长春130051
  • 收稿日期:2014-07-08 修回日期:2015-03-11 出版日期:2015-05-15 发布日期:2015-05-25
  • 基金资助:

    国家自然科学基金（71331008, 71101150）

Prediction Method for Imaging Task Schedulability of Earth Observation Network

LIU Song ，BAI Guo qing，CHEN Ying wu   

  1. 1. College of Information System and Management，National University of Defense Technology，Changsha 410073，China；
     2. The Military Representative Office Stationed in Changchun，Changchun 130051，China
  • Received:2014-07-08 Revised:2015-03-11 Online:2015-05-15 Published:2015-05-25

PDF

865

摘要：

为了能够快速、合理地分配成像任务，充分发挥对地观测网络的观测效能，对成像任务可调度性预测问题进行了研究，提出一种由协同任务分配组件、任务调度组件、特征提取组件以及任务可调度性预测组件所构成的组件化求解架构。在成像卫星经典调度模型的基础上，提取成像任务特征，并采用变隐含层节点的反向传播（BP）神经网络集成技术求解成像任务可调度性问题。仿真结果表明，集成BP神经网络的平均预测准确度可以达到85%以上。

关键词: 对地观测网络, 任务可调度性, 预测, 神经网络集成, BP神经网络

Abstract:

In order to achieve assigning imaging tasks quickly and efficiently in the earth observation network, a novel component-based solution structure. Composed of task coordinated allocator, task scheduler, feature extractor and schedulability predictor is proposed. Based on the classic imaging satellite scheduling model, the features of imaging tasks are extracted, and the imaging task scheduling prediction problem is solved by using the BP neural network ensemble technique for variable hidden layer nodes. Simulation results demonstrate that the back propagation (BP) neural network ensemble used in this paper for a single imaging satellite can reach daily schedulability prediction accuracy more than 85%.

Key words: Earth observation network, Task schedulability, Predict, Neural network ensemble, BP neural network

中图分类号: