摘要
由于地磁图适配区的选择是影响地磁导航定位精度的重要因素,因此提出一种基于主成分分析法(PCA)和GA-BP神经网络相结合的地磁背景场适配/非适配区自动识别和分类的方法。首先利用PCA对地磁特征参数进行分析,选择出独立的、并且包含主成分的特征参量,其次构建GA-BP神经网络模型,建立地磁特征参数和匹配性能的对应关系,从而实现适配/非适配区的划分。通过多次仿真试验,证明了采用该方法能够选择出较好的适配区域,提高地磁导航定位精度。
The selection of suitable matching area of geomagnetic map is important for ensuring the positioning accuracy of geomagnetic navigation. This paper puts forward a method for the automatic recognition and classification of the suitable and unsuitable matching areas of geomagnetic background field based on Principal Component Analysis（ PCA） and GA-BP neural network. To select independent characteristic parameters containing the main components,PCA is used to analyze the geomagnetic characteristic parameters. Then,the GA-BP neural network model is constructed,and the correspondence between the geomagnetic characteristic parameters and matching performance is established,so as to realize the recognition and classification of suitable and unsuitable matching areas. Simulation results show that this method can efficiently find out a more effective matching area,and improve the positioning accuracy of geomagnetic navigation.
作者
王晨阳
WANG Chen-yang(School of Automation, Nanjing University of Science and Technology, Nanjing 210094, China)
出处
《电光与控制》
北大核心
2018年第6期110-114,共5页
Electronics Optics & Control
关键词
地磁导航
特征参数
主成分分析
GA-BP神经网络
适配区选择
geomagnetic navigation
characteristic parameter
principal component analysis
GA-BP neural network
matching area selection
