指挥控制与仿真2026，Vol.48Issue(3)：18-29,12.DOI:10.3969/j.issn.1673-3819.2026.03.003
基于GraphRAG的无人机集群作战效能评估
Knowledge Graph-based evaluation of uav swarm combat effectiveness
摘要
Abstract
To address the current lack of a pathway from deduction simulation to evaluation in the assessment of UAV swarm operational effectiveness,a GraphRAG-based evaluation method was proposed,which fully utilizes deduction simulation data to reflect the operational effectiveness of UAV swarms in a realistic and objective manner.First,a knowledge graph was con-structed using data from UAV swarm deduction simulations,such as attack,detection,and operational effect data.Subse-quently,a Graph Retrieval-Augmented Generation(GraphRAG)system based on evaluation indicators was developed using the knowledge graph,combined with a large language model to intelligently generate a reasonable evaluation indicator sys-tem.Finally,the information entropy method was employed to achieve the operational effectiveness evaluation of UAV swarms.By constructing a UAV swarm ground attack operational scenario,the effectiveness of the proposed method was veri-fied,opening a new methodological pathway for system effectiveness evaluation.关键词
无人机集群/知识图谱/图检索增强生成/信息熵法/效能评估Key words
UAV swarm/knowledge graphs/GraphRAG/information entropy method/effectiveness evaluation分类
军事科技引用本文复制引用
李书杰,杨成伟,张鹏飞,杨昭..基于GraphRAG的无人机集群作战效能评估[J].指挥控制与仿真,2026,48(3):18-29,12.基金项目
国家自然科学基金(62103048) （62103048）
