论文部分内容阅读
为了适应低成本快速响应卫星的发展趋势,越来越多的卫星采用了MEMS陀螺系统与信息融合方法相结合的设计方案。为提高传统的基于自回归滑动平均(Auto-Regressive and Moving Average,ARMA)模型和支持度的信息融合方法的精度,抑制随机误差趋势项对卫星稳定控制的不利影响,提出了一种基于支持向量回归机(Support Vector Regression,SVR)的预测补偿算法。该算法对各个MEMS陀螺输出数据进行滤波,并提取相应的随机误差趋势项,通过相空间重构获得训练样本并进行SVR建模,用以实时补偿。然后使用低成本商用器件搭建了MEMS陀螺系统,并在单轴气浮转台上进行了实验。实验结果表明,预测补偿算法使得MEMS陀螺系统输出数据的方差降为原先的31.39%,融合精度得到了显著的提高。
In order to adapt to the development trend of low cost and quick response satellite, more and more satellites adopt the design scheme of MEMS gyroscope system combined with information fusion method. In order to improve the accuracy of the traditional information fusion method based on the Auto-Regressive and Moving Average (ARMA) model and support, and to suppress the adverse effect of the random error trend items on the satellite stability control, a new method based on support vector Predictive compensation algorithm for Support Vector Regression (SVR). The algorithm filters each MEMS gyro output data and extracts the corresponding random error trend items. The phase space reconstruction obtains the training samples and performs SVR modeling to compensate in real time. Then the MEMS gyroscope system was built using low cost commercial devices and the experiment was carried out on a single axis air floating platform. The experimental results show that the prediction compensation algorithm makes the variance of the output data of the MEMS gyroscope system reduced to 31.39% of the original, and the fusion accuracy has been significantly improved.