基于核Fisher鉴别分析的RVM战斗机空战效能评估  

RVM Effectiveness Evaluation of Air Combat Based on Kernel Fisher Discriminant Analysis

作　　者：任水利[1] 甘旭升[2] 丁黎颖 杨国洲 REN Shui-li;GAN Xu-sheng;DING Li-ying;YANG Guo-zhou(Xijing University,Xi’an 710123,China;School of Air Traffic Control and Navigation,Air Force Engineering University,Xi’an 710051,China)

机构地区：[1]西京学院,西安710123 [2]空军工程大学空管领航学院,西安710051

出　　处：《火力与指挥控制》2019年第10期143-147,152,共6页Fire Control & Command Control

摘　　要：为提高战斗机空战效能的评估能力,提出一种基于KFDA特征提取的RVM效能评估方法。该方法在系统分析基础上,引入战斗机空战效能评估指标体系;采用KFDA提取评估指标数据中的高阶非线性特征;凭借RVM在非线性系统建模方面的优势,构建所提取特征的战斗机空战效能评估模型,进而完成效能评估。实例分析表明,该方法对战斗机空战效能评估的均方根误差达到0.0094,评估结果相对于其他方法更为准确,过程也更为科学合理,从而验证了其用于战斗机空战效能评估是可行和有效的。To improve the ability of effectiveness evaluation of warcraft’s air combat,a Relevance Vector Machine (RVM ) effectiveness evaluation method is proposed based on Kernel Fisher Discriminant Analysis (KFDA )feature extraction. Firstly, on the basis of system analysis, the index system of effectiveness evaluation of fighter air combat is introduced. Then, KFDA is used to extract higher-order nonlinear features from evaluation index data. Finally, the advantages of RVM in nonlinear system modeling are applied to establish the effectiveness evaluation model of warcraft’s air combat based on extracted features and complete the effectiveness evaluation. The actual example shows that, the root mean square error obtained by proposed method can achieve 0.009 4 for effectiveness evaluation of warcraft’s air combat. It has a scientific and reasonable process and an accurate evaluation result. This verifies its feasibility and effectiveness in effectiveness evaluation of warcraft’s air combat.

关 键 词：相关向量机 核FISHER鉴别分析 特征提取 战斗机 效能评估 

分 类 号：TJ85[兵器科学与技术—武器系统与运用工程] V271.4[航空宇航科学与技术—飞行器设计]