基于SSA-SVM的空中目标意图识别方法  

Intention Recognition Method of Air Target Based on SSA-SVM

作　　者：吴广宇[1,2] 史红权 邱楚楚[1] WU Guangyu;SHI Hongquan;QIU Chuchu(Dalian Naval Academy,Dalian 116018;No.91991 Troops of PLA,Zhoushan 316041)

机构地区：[1]海军大连舰艇学院,大连116018 [2]中国人民解放军91991部队,舟山316041

出　　处：《舰船电子工程》2022年第3期29-34,共6页Ship Electronic Engineering

摘　　要：针对传统的网格搜索算法(GSA)和粒子群算法(PSO)分别优化的支持向量机(SVM)无法同时满足舰船对空中目标意图识别中准确性和快速性的需要而导致识别效果较差的问题,提出了一种基于麻雀搜索算法(SSA)优化支持向量机的空中目标意图识别方法,建立了基于SSA-SVM的空中目标意图识别模型,并通过仿真实验与上述两种算法优化的支持向量机模型进行效果对比。结果表明,相比GSA-SVM和PSO-SVM,SSA-SVM同时具有较高的识别准确率和较短的识别运算时间。因此SSA-SVM能够既准确又快速地识别空中目标意图,具有更好的识别效果,验证了该方法的有效性。Aiming at the problem that the support vector machine(SVM)optimized by the traditional grid search algorithm(GSA)and particle swarm optimization(PSO)cannot simultaneously meet the requirements of accuracy and quickness in intention recognition of air target for ship,which results in the poor recognition effect.This paper proposes an intention recognition method of air target based on the support vector optimized by sparrow search algorithm(SSA).The air target intention recognition model based on SSA-SVM is established.And its effect is compared with the effect of support vector machine model optimized by the above two algorithms through simulation experiments.As a result,compared with GSA-SVM and PSO-SVM,SSA-SVM simultaneously has both higher recognition accuracy and shorter recognition operation time.Therefore,SSA-SVM can recognize the intention of air target accurately and quickly and obtain the better recognition effect,which verifies the effectiveness of the method.

关 键 词：支持向量机 网格搜索算法 粒子群算法 麻雀搜索算法 空中目标意图识别 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] E911[自动化与计算机技术—计算机科学与技术]