梯度RBF神经网络在MEMS陀螺仪随机漂移建模中的应用

Application of gradient radial basis function network in the modeling of MEMS gyro's random drift

为了提高使用精度,研究了某型号MEMS陀螺仪的随机漂移模型。采用游程检验法分析了该陀螺仪随机漂移数据的平稳性,并根据该漂移为均值非平稳、方差平稳的随机过程的结论,采用梯度径向基(RBF)神经网络对漂移数据进行了建模。实验结果表明:相比经典RBF网络模型而言,这种方法建立的模型能更好地描述MEMS陀螺仪的漂移特;相对于季节时间序列模型而言,其补偿效果提高了大约15%。

In this paper the random dritf of a micro-electro-mechanical system （MEMS） gyro is analyzed and modeled in order to improve its performance. The analysis, which is based on run test, shows that the drift is not a weak stationary random process. Thus, a model based on gradient radial basis function neural network is applied to deal with the non-stationary process. The experiments show that the model is suitable for the random drift.