捷联陀螺漂移误差系数预测方法研究

A method for predicting the drift error coefficients of strap-down gyroscope

针对目前通用的均值估计法和AR(P)模型直接预测法对捷联陀螺漂移误差系数预测精度不高的情况,对AR(P)模型直接预测法进行了改进,建立了基于误差系数漂移量的AR(P)模型,首先预测漂移量,进而预测出漂移误差系数值。通过仿真实例对3种方法的预测效果进行比较分析,结果证明本文提出的预测方法优于目前通用的两种预测方法。

Considering the low precision of such general methods as average estimation and AR（P） model direct prediction in predicting the drift error coefficients of strapdown gyroscope, an improved AR（P） model is presented in this paper. The AR（P） model is built based on the driftance of error coefficients. First, predict the driftance, and then predict the drift error coefficients. The simulation indicates that the new prediction method proposed is better than the two general methods currently used, and this method possesses a good prospect .