Modeling random gyro drift by time series neural networks and by traditional method

Abstract

This paper presents modeling random gyro drift rate by traditional time series method , and makes compensation for gyro drift by Kalman filter, and proposes the modeling and forecasting method by neural networks for strapdown gyro based on time series analysis, and makes a research for random drift rate of gyro applied for strapdown inertial navigation systems, comparison between the results of by Kalman filter based traditional time series method and by time series neural networks is presented.