联合SVM和HMM的水上/水下导航场景感知模型构建  

Hybrid SVM and HMM based navigation context awareness models for overwater and underwater mixed scene

作　　者：朱锋 罗科干[2] 陈惟杰 刘万科 张小红[1,2] ZHU Feng;LUO Kegan;CHEN Weijie;LIU Wanke;ZHANG Xiaohong(Hubei Luojia Laboratory,Wuhan 430079,China;School of Geodesy and Geomatics,Wuhan University,Wuhan 430079,China)

机构地区：[1]湖北珞珈实验室,湖北武汉430079 [2]武汉大学测绘学院,湖北武汉430079

出　　处：《测绘学报》2023年第5期738-747,共10页Acta Geodaetica et Cartographica Sinica

基　　金：国家重点研发计划(2020YFB0505803);国家自然科学基金(42104021);湖北省科技重大项目(2021AAA010);湖北珞珈实验室专项(220100005)。

摘　　要：导航场景感知是智能化PNT的重要特征,更是实现多场景无缝导航定位的基础。本文聚焦水上/水下导航场景,考虑电磁波的衰减程度差异将其细分为水上、浅水、深水3类场景,利用支持向量机(support vector machine,SVM)进行场景分类与识别,在此基础上,引入隐马尔可夫模型(hidden Markov model,HMM)表达导航场景切换,进一步提升场景识别可靠性。本文分别构建了基于结果联合(SVM-HMM1)及基于概率联合(SVM-HMM2)的水上/水下导航场景感知模型。实测分析表明,两种模型能够实现高精度场景感知,SVM-HMM1与SVM-HMM2识别准确率分别为91.36%与95.11%;与单一的HMM和SVM模型相比,联合模型在结果分类与识别上更为稳定,准确率提升约为0.95%~8.46%。Navigation context awareness is not only an important feature of intelligent PNT(positioning,navigation and timing),but also the basis for realizing multi-scene seamless navigation and positioning.This paper focuses on the overwater and underwater scene,which is divided into three kinds of subdivide scenes:overwater,shallow water and deep water according to the change of GNSS signal characteristics,using support vector machine(SVM)for scene classification and recognition.On this basis,hidden Markov model(HMM)is introduced to express navigation scene switching to further improve the reliability of context awareness.This paper constructs two kinds of context awareness models based on result combination(SVM-HMM1)and probability combination(SVM-HMM2).The recognition accuracy of SVM-HMM1 and SVM-HMM2 are 91.36%and 95.11%,respectively.Compared with HMM and SVM,the combined models are more stable in result classification and recognition,and the accuracies are improved by about 0.95%~8.46%.

关 键 词：智能PNT 导航场景感知 水上/水下导航场景 支持向量机 隐马尔可夫模型 

分 类 号：P228[天文地球—大地测量学与测量工程]