摘要
GPS高程转换是GPS高程应用的关键问题之一 ,试图应用神经元网络方法来进行GPS高程的转换 ,并在网络的拓扑结构与网络结构的改进等方面作了分析 ,通过GPS高程转换的二次曲面法和神经元网络法进行了比较试验 ,结果表明了神经元网络转换GPS高程的可行性和可靠性 ,对于应用神经网络转换GPS高程具有实际的指导意义。
Comparing with fitting method that simulates Geoid using quadric polynomial for conversion of GPS height, this paper further shows the feasibility of GPS height conversion by using artificial neural network. Theoretical and practical computations have shown that artificial neural network is suitable for GPS height conversion. At the same time, some problems, such as improving structure of BP network model are deeply investigated. Some conclusions the paper drown has the theoretical guiding function for the GPS trasformation using neural network.
出处
《全球定位系统》
2004年第2期24-28,共5页
Gnss World of China
基金
国家自然科学基金资助项目 (4 0 1710 82 1)
