一种空中编队队形智能识别方法  

An Intelligent Recognition Method of Aircraft Formation

作　　者：梁复台 周焰[1] 张怀念 张勇[3] 赵小瑞 LIANG Futai;ZHOU Yan;ZHANG Huainian;ZHANG Yong;ZHAO Xiaorui(Department of Early Warning Intelligence,Air Force Early-Warning Academy,Wuhan 430014,China;Unit 31121,PLA,Nanjing 210042,China;Department of Air defense Early Warning Command,Air Force Early-Warning Academy,Wuhan 430014,China)

机构地区：[1]空军预警学院预警情报系,湖北武汉430014 [2]中国人民解放军31121部队,江苏南京210042 [3]空军预警学院防空预警指挥系,湖北武汉430014

出　　处：《无线电工程》2023年第7期1604-1611,共8页Radio Engineering

基　　金：军队重大科研课题(JY2020A020)。

摘　　要：针对传统空中编队队形识别方法人工干预多、模板设计复杂等问题,提出一种基于态势编码和机器学习的空中编队队形智能识别方法。提出了一种适用于机器学习算法的空中编队态势编码方法,对不同空中编队队形进行统一描述;以空中编队态势编码为基础,构建基于机器学习的编队队形识别模型;仿真编队数据并划分数据集,对模型进行训练。实验结果表明,该方法在小样本条件下识别准确率可达95.5%以上,在样本数据量较大时可达99%,可实现端到端的识别。For the problems of manual intervention and complex template design in the traditional recognition method of aircraft formation based on the template,an intelligent recognition method of aircraft formation based on situation coding and machine learning is proposed.A situation coding method of aircraft formation suitable for machine learning algorithms is designed to uniformly describe different aircraft formations.Based on aircraft formation situation coding,an aircraft formation recognition model based on machine learning is designed.The simulation data of aircraft formation is formed data set and used to train the model.The experimental results show that the recognition accuracy of proposed method can reach more than 95.5%under the condition of small samples,and can reach 99%when the sample data is large.The method can realize end-to-end recognition.

关 键 词：态势编码 空中编队 队形识别 机器学习 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]