电子学报 ›› 2012, Vol. 40 ›› Issue (5): 901-906.DOI: 10.3969/j.issn.0372-2112.2012.05.007

• 学术论文 • 上一篇    下一篇

基于Elman-AdaBoost强预测器的目标威胁评估模型及算法

王改革1,2, 郭立红1, 段红3, 刘逻1,2, 王鹤淇1   

  1. 1. 中国科学院长春光学精密机械与物理研究所, 吉林长春 130033;2. 中国科学院研究生院, 北京 100039;3. 东北师范大学计算机科学与信息技术学院, 吉林长春 130117
  • 收稿日期:2011-07-12 修回日期:2011-12-16 出版日期:2012-05-25
    • 通讯作者:
    • 郭立红
    • 基金资助:
    • 激光与物质相互作用国家重点实验室研究基金 (No.SKLLIM0902-01)

The Model and Algorithm for the Target Threat Assessment Based on Elman-AdaBoost Strong Predictor

WANG Gai-ge1,2, GUO Li-hong1, DUAN Hong3, LIU Luo1,2, WANG He-qi1   

  1. 1. Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences. Changchun, Jilin 130033, China;2. Graduate School of Chinese Academy of Sciences, Beijing 100039, China;3. School of Computer Science and Information Technology, Northeast Normal University, Changchun, Jilin 130117, China
  • Received:2011-07-12 Revised:2011-12-16 Online:2012-05-25 Published:2012-05-25

摘要: 目标威胁评估是协同目标攻击中的关键问题.为提高空战目标威胁评估的准确性和实用性,建立了Elman-AdaBoost强预测器目标威胁评估模型及算法.首先,介绍了Elman-AdaBoost强预测器;其次,建立了Elman-AdaBoost强预测器目标威胁评估模型;最后,提出了基于Elman-AdaBoost强预测器目标威胁评估模型的算法.采集75组数据用于实验,其中60组作为训练集,15组作为测试集.分别选择Elman网络隐层节点数L=7,11,14,18和弱预测器数目K=6,10,16,20进行实验,结果表明,Elman-AdaBoost强预测器算法预测误差远小于弱预测器且在L=7和K=6时误差达到最小.Elman-AdaBoost强预测器目标威胁评估模型和算法具有很好的预测能力,可以快速、准确地完成作战目标威胁评估.

关键词: 目标威胁评估, 模型, 算法, Elman-AdaBoost

Abstract: Target threat assessment is the key issue in the collaborative multi-target attack.To improve the accuracy and usefulness of target threat assessment in the aerial combat,a target threat assessment model and algorithm based on Elman-AdaBoost strong predictor is proposed.Firstly,Elman-AdaBoost strong predictor is introduced;secondly,a target threat assessment model based on Elman-AdaBoost strong predictor is established;at last,an algorithm is described.There are 75 data sets culled for the simulation experiments,in which 60 sets are considered as training set,and the other 15 are testing sets.The number of hidden layer nodes of Elman network and weak predictors is selected L=7,11,14,18 and K=6,10,16,20 respectively for experiment and results show that,the prediction error for Elman-AdaBoost strong predictor algorithm is much smaller than the weak predictor and the error reaches the minimum when L=7 and K=6.The model and algorithm have good predictive ability,so it can quickly and accurately complete target threat assessment.

Key words: target threat assessment, model, algorithm, Elman-adaptive boosting

中图分类号: