摘要
针对卫星激光测距望远镜指向误差问题,提出一种基于BP神经网络的建模方法。使用武汉流动卫星激光测距站的三组恒星观测数据,对比了BP神经网络模型、转台模型、球谐函数模型和基本参数模型的中误差。研究结果表明,BP神经网络算法可应用于指向误差建模,且建模精度优于其他三种模型。
A modeling method based on the BP neural network is proposed for correcting pointing error of the satellite laser ranging telescope. This experiment makes use of the three groups of stars data observed in Wuhan mobile satellite laser ranging station to calculate and compare root mean square errors by respectively establishing BP neural network model, mount model, spherical harmonics model and basic parameters model. The results show that BP neural network can be applied to pointing error modeling, and modeling accuracy is higher than other three models.
出处
《大地测量与地球动力学》
CSCD
北大核心
2013年第5期150-153,共4页
Journal of Geodesy and Geodynamics
基金
中国地震局地震研究所所长基金(IS201156086)
关键词
指向误差
转台模型
BP神经网络
中误差
卫星激光测距
pointing error
mount model
BP neural network
root mean square error
Satellite Laser Ranging（SLR）
