平台动基座标定中的参数辨识研究

给出了某型平台动基座标定的参数辨识的一个工程实例。针对阵风干扰对标定精度影响很大的情况，设计了小波滤波加分组拟合和加权平均算法。在使用小波滤波器对数据滤波后，采用分组拟合加权平均算法辨识漂移角速度，在此基础上解算漂移参数。从试验数据计算值可以看出，该方法对扰动和随机误差的消除效果明显，优于最小二乘法。

Parameter Identifying in Calibration of Platform on Shake Base

This paper presents an example of parameter identifying in calibration of gyro-platform on shake base. In view of the great influence of the error caused by flurry, the new algorithm of parameter identifying, which synthetically uses wavelet filter, grouping regression and weighting average, is designed. First, the output data of platform are filtered in advance by wavelet filter. Then, the drift angle velocity is obtained by means of grouping regression and weighting average from the filtered data. The drift coefficients are calculated from the drift angle velocity. It is proved through experiment that the new algorithm can efficiently reduce the interfere error and random error. Its effect is better than that of LS.