基于多维数据关联的航天观测任务衍生技术研究

打开文本图片集

引用格式：，，，等.基于多维数据关联的航天观测任务衍生技术研究[J].指挥控制与仿真，2025，47（6）：69-75.TANJ，GUOQ，ZHANGXL，etal.SpaceobservationtaskderivationbasednultidmensioaldataasocationJ].ComandCotrol&Simulation，2025，47（6） ：69-75.

中图分类号：TP311 文献标志码：A DOI：10. 3969/j. issn.1673-3819.2025.06.010

Space observation task derivation based on multidimensional data association

TAN Juan¹，GUO Qi2 ，ZHANG Xueliang³，ZHANG Wenbao² （1.Beijing Institute of Tracking and Communication Technology，Beijing 1Ooo94，China; 2.Key Laboratory of Aerospace Information Applications of CETC，Shijiazhuang O5oo81，China; 3.The 54u Research Institute of CETC，Shijiazhuang O5Oo81，China）

Abstract：With the increasingdemandforremotesensing observation fromusers in various industries，thecurentapplication modeof "pasive response touserdemand"forspaceobservationresources is becoming more andmore limited.Therefore，research onspaceobservation mision derivation technology basedonmultidimensional dataasociation iscariedout. Firstly，thediferentiatedspaceresourcesaremodeled toaccuratelydescribethecapabilityaributesofvariousspaceresourcesandspaceobservationmision knowledge.Secondly，forallkindsof explicitorimplicitspaceobservationrequirements，he method basedonlarge language modelisused tounderstandtherequirementsand extractthe task elements; Then，amulti-semantic inference network isconstructedtoimplement thederivedreasoning of aerospace observationtasks. Afterthat，thechangerulesofthehistoricaltaskresourcesupplyanddemandrelationshipof typicalusersare minedtoautomaticallyrecommend spaceresources.Finally，thederivationprocessofspaceobservationmision inemergencydisasterrelief scenarioistakenasanexample to verifythemethod.Theexampleanalysis shows thatthismethodcannotonly esure the accuracy，but alsoeffectivelyimprove theoperation eficiencyanduseeficiencyoftheaerospaceobservationsystem.

Keywords：space observation；task derivation；multi-semantic inference network

随着在轨卫星数量和能力的不断提高，用户的观测需求越来越旺盛，航天观测在农业生产、应急救灾等领域内的作用越来越重要。（剩余9672字）