空天海地一体化观测网络的任务管理探讨与展望

Discussion and Prospect of Task Management of Space-Air-SeaLand Integrated Observation Network

观测网络的网络拓扑动态性、资源异构性及其系统复杂性给空天海地一体化观测网络的任务管理带来了巨大的挑战。本文首先研究探讨面向一体化观测网络及其任务管理的主要内容、关键技术及其面临的挑战。为了应对观测网络面临的问题,本文将基于ACP的平行智能方法引入到观测网络任务管理领域,提出平行观测网络的概念与架构。平行观测网络首先通过人工观测网络系统来模拟实际观测网络系统;然后,进行大量不同观测场景及事件的计算实验进行训练与评估;最后,通过平行执行来滚动在线优化人工和实际观测网络系统,从而实现对复杂观测网络的描述智能、预测智能和引导智能。因此,可以实现根据观测环境及任务需求快速灵活地组网,提高网络的灵活性及任务管理效率。平行观测网络将为国家网络数据安全、能源与资源开发、环境监测、灾害预报预警等研究提供有效支撑。

The topological dynamics,resource heterogeneity and system complexity of observation networks bring great challenges to the task management of space-air-sea-land integrated observation networks.This paper first studies the main content,key technologies and challenge of integrated observation networks and its task management.To deal with the problem faced by the observation network,this paper proposes to introduce the ACP-based parallel intelligent method into the field of observation network task management,and proposes the concept and architecture of parallel observation network.The parallel observation networks simulate the heterogeneous and complex actual observation network by using artificial observation network system.The training and evaluation of various observation scenarios and event models are carried out through computational experiments.Then,perform a large number of computational experiments on different observation scenarios and events for training and evaluation.Finally,the observation network system is optimized by parallel execution,the descriptive intelligence,predictive intelligence and prescriptive intelligence of complex observation network are achieved.So that the system can quickly and flexibly construct networks according to the observation environment and task requirements to improve the flexibility of network and efficiency of task management.The proposed parallel observation networks will provide effective support for national network data security,energy and resource development,environmental monitoring,disaster warning and forecasting and other areas.