陀螺随机漂移的神经网络预报方法研究

针对惯性导航系统中陀螺仪的漂移特性，在时间序列分析及神经网络理论的基础上提出了一种时间序列神经网络结构，并采用此种网络模型对某捷联惯导系统中所用的陀螺仪漂移数据进行了预报。预测结果表明，这种预测方法对于陀螺漂移建模及预报是可行的。

Forecasting Random Drift Rate for Strapdown Gyro by Neural Networks

This paper proposes a new net of the time series and neural networks for the drift of gyroscopes used in the strapdown inertial systems based on time series analysis and net networks theory, and makes a forecasting for random drift rate of gyro applied for strapdown inertial navigation systems. The results show this forecasting method is effective for modeling and forecasting for gyroscopic drift.