基于信号分解的防御弹制导律辨识方法  

A method of guidance law identification for defense missile based on signal decomposition

作　　者：王晓芳[1] 张楠 Wang Xiaofang;Zhang Nan(School of Aerospace Engineering,Beijing Institute of Technology,Beijing 100081,China)

机构地区：[1]北京理工大学宇航学院,北京100081

出　　处：《战术导弹技术》2024年第1期95-104,共10页Tactical Missile Technology

摘　　要：针对飞机对敌方防御弹制导律辨识的问题,提出一种对相对运动信息进行经验模态分解并采用支持向量机分类的辨识方法。建立了防御弹和飞机的相对运动模型,并根据典型作战场景构建了防御弹可能采取的制导律模型集。针对雷达测量的防御弹与飞机间的视线角速度及距离变化率信号,采用经验模态分解算法对其进行信号分解,提取并构建不同制导律的特征能量带。建立“一对一”多分类支持向量机模型对不同制导律的特征能量带样本进行分类,从而实现制导律的辨识。仿真结果表明,本研究提出的基于经验模态分解算法和支持向量机模型的制导律辨识方法能够以较高精度实现对样本集中制导律的辨识。Aiming at the problem that aircraft can identify the guidance law of the enemy defense missile,an identification method based on empirical mode decomposition of relative motion information and classification using support vector machine is proposed.The relative motion model of aircraft and defense missile is established.The possible guidance law model set of defense missile is constructed according to typical combat scenarios.The empirical mode decomposition algorithm is used for decomposing the LOS angular velocity and range rate signal measured by the radar between the defense missile and the aircraft.The characteristic energy matrix of different guidance laws is constructed."One-to-One"multi-class support vector machine model is established to classify the sample of characteristic energy matrix of different guidance laws,to realize the identification of guidance laws.The simulation results show that the proposed guidance law identification method based on empirical mode decomposition algorithm and support vector machine model can identify the guidance law in sample set with high accuracy.

关 键 词：制导律辨识 机器学习 比例导引律 相对运动信息 经验模态分解 能量特征 多分类支持向量机 

分 类 号：TJ765[兵器科学与技术—武器系统与运用工程]