[1] Hu Wusheng, Sha Yuejin,. Converting GPS height by a new methodbased on neural networks [J]. Journal of Southeast University (English Edition), 2003, 19 (1): 53-57. [doi:10.3969/j.issn.1003-7985.2003.01.013]

Copy

Converting GPS height by a new methodbased on neural networks()

基于神经网络的转换GPS高程的新方法

Journal of Southeast University (English Edition)[ISSN:1003-7985/CN:32-1325/N]

Volumn:

19

Issue:

2003 1

Page:

53-57

Research Field:

Surveying and Mapping and Navigation

Publishing date:

2003-03-30

Info

Title:

Converting GPS height by a new methodbased on neural networks

基于神经网络的转换GPS高程的新方法

Author(s):

Hu Wusheng, Sha Yuejin

College of Transportation, Southeast University, Nanjing 210096, China

胡伍生, 沙月进

东南大学交通学院测绘工程系, 南京 210096

Keywords:

neural networks;  algorithm of BP;  GPS height;  CF&;  NNM

神经网络;  BP算法;  GPS高程;  CF&;  NNM方法

PACS:

P228

DOI:

10.3969/j.issn.1003-7985.2003.01.013

Abstract:

The adjusted GPS height is the height above the surface of the WGS-84 ellipsoid. It is necessary to convert a GPS height into a normal height in engineering. The conicoid fitting method(CFM)and the neural networks method(NNM)are used for this purpose, but each of them has its advantages and disadvantages. After studying these two methods, a new method(abbr. CF&NNM)is conceived. The procedure of the CF&NNM is introduced. A practical engineering example is used to study these three different methods. The results by the three methods are listed. The CF&NNM method can produce better results than either the CFM or the NNM in deriving normal height from GPS height. The theory of the CF&NNM method is analyzed.

GPS高程是相对于WGS-84椭球体的大地高, 因此, 在工程应用中, GPS高程需要转换为正常高.转换GPS高程通常采用二次曲面拟合法(CFM)和神经网络方法(NNM), 但这2种方法各有优缺点.在研究了这2种方法之后, 提出了一种转换GPS高程的新方法, 该方法综合了上述2种方法的优点, 故取名为“CF&NNM”方法.介绍了CF&NNM方法的思路和计算过程.通过一个工程实例, 列出了上述3种方法的数据处理结果, 新方法效果最好.对CF&NNM方法进行了理论分析.

References:

[1] Liu Dajie, Shi Yimin, Guo Jingjun. The theory of GPS and its application[M]. Shanghai: Tongji University Press, 1996.224-229.(in Chinese)
[2] Yuan Zengren. Artificial neural networks and their applications [M]. Beijing: Tsinghua University Press, 1999. 66-130.(in Chinese)
[3] Yang Mingqing, Jin Fan, Zhu Dacheng, et al. Conversion of GPS height by artificial neural network method [J]. Acta Geodaetica et Cartogra Phica Sinica, 1999, 28(4):301-307.(in Chinese)
[4] Hu Wusheng. The research of the theory and the method and the application of precise GPS height survey[D]. Nanjing: Hohai University, 2001.33-78.(in Chinese)
[5] Leonard J, Kramer M A. Improvement of the back propagation algorithm for training neural networks [J]. Computers Chem Engng, 1990, 14(3):337-341.

Memo

Memo:

Biography: Hu Wusheng(1965—), male, doctor, professor, ws.hu@
jlonline.com.

Common functions