面向星群的遥感影像智能服务关键问题

Key problems of remote sensing images intelligent service for constellation

全天时、全天候和全球的遥感信息实时智能服务是对地观测系统建设的目标。近几年来,随着我国高分专项和商业卫星的发展,在轨卫星数量急剧增加,对地观测能力得到极大增强,使得传统的单星和星座卫星系统的运控、接收、处理和应用服务模式面临严峻挑战,亟须统筹规划卫星应用各环节资源,充分发挥多星协同优势,构建统一的遥感影像实时智能服务体系和系统。本文针对遥感星群卫星体系特点和对地观测用户需求特征,开展面向星群的遥感影像智能服务关键问题研究。提出了面向任务的全球多尺度语义描述网格,统筹全球动静态的任务语义描述,在此基础上重点分析了面向星群的自主任务管理、精准动态规划和协同智能处理等关键技术问题,形成集任务描述、任务管控、任务规划、在轨处理、终端分发一体化的星群智能服务技术体系。通过充分发挥星群协同的优势,结合在轨处理和人工智能技术来降低各环节时间延迟,提高数据处理精度,从而实现完全自动化和智能化的近实时星群智能服务,为对地观测的全天时、全天候快速高效智能服务奠定基础。

The all-day,all-weather and global real-time intelligent service of remote sensing information is the goal of earth observation system construction.In recent years,with the development of domestic high-scoring special projects and commercial satellites,the number of on orbit satellites has increased dramatically,and the earth observation capability has been greatly enhanced,which makes the operation control,reception,processing and application service mode of the traditional single satellite and constellation satellite systems face severe challenges.It is urgent to plan the resources of all aspects of satellite application,give full play to the advantages of multi-satellite collaboration,and build a unified remote sensing images real-time intelligent service system.According to the characteristics of remote sensing constellation satellite system and the demand characteristics of earth observation users,this paper studies the key problems of remote sensing image intelligent service for constellation.This paper proposed a task oriented global multi-scale semantic description grid to coordinate the global dynamic and static task semantic description.On this basis,it focuses on the key technical problems such as autonomous task management,intelligent planning and multi satellite collaborative intelligent service.Formed a constellation intelligent service technology system integrating mission control,mission planning,on-orbit processing,data download and distribution.By giving full play to the advantages of constellation collaboration,combining on-orbit processing and artificial intelligence technology to reduce the time delay of each link and improve the data processing accuracy.So as to achieve fully automated and intelligent near real-time constellation intelligent services,and lay the foundation for all-day,all-weather,fast and efficient intelligent services for earth observation.