灰色模型与神经元网络结合的飞行目标预测法  

Flight Target Prediction Method Combined with Grey Model and Neural Network

作　　者：潘峰 熊培蕾 葛橹漠 PAN Feng;XIONG Peilei;GE Lumo(Unit 31006 of PNA,Beijing 100036,China;Unit 61886 of PLA,Beijing 100085,China;The 28th Research Institute of China Electronics Technology Group Corporation,Nanjing 210007,China)

机构地区：[1]解放军31006部队 [2]解放军61886部队 [3]中国电子科技集团公司第二十八研究所

出　　处：《指挥信息系统与技术》2018年第6期66-69,共4页Command Information System and Technology

摘　　要：利用机动飞行目标轨迹可预测飞行目标实时位置,针对灰色模型对快速飞行目标轨迹预测不准确的问题,引入神经元理论判断飞行目标状态,通过机器学习掌握飞行目标转弯模型,从而更准确地对飞行目标转弯进行预测。首先,分析了灰色模型预测方法的工作原理和特点;然后,借助反向传播(BP)神经元模型可判断飞行特征的优势,将灰色模型与神经元网络模型进行了结合;最后,通过仿真试验表明该飞行目标预测法可对飞行目标进行精准监视和控制。The real-time location of flight target can be predicted by the maneuvering target trajectory.Aimed at the inaccuracy problem of the fast flying target trajectory prediction by the grey model,the neural theory is introduced to judge the flight target mode,the flight target turning model is controlled by the machine learning,and the turning situation of flight target is predicted accurately.Firstly,the working principle and characteristics of grey model prediction method are analyzed.Then,the grey model and the neural network model are combined through the advantage of the back propagation(BP)neuralmodel to judge the flight characteristics.Finally,the simulation result shows that this method can accurately monitor and control the flight target.

关 键 词：航迹预测 机动飞行目标 BP神经元网络 灰色模型 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]