改进小波阈值法与极限学习机在MEMS陀螺误差补偿中的应用

MEMS陀螺随机误差是影响其精度的主要因素之一。针对MEMS陀螺随机误差的问题，提出一种基于改进的阈值函数的小波去噪结合极限学习机算法建模的补偿方法。通过改进小波阈值法提高去噪效果，然后由极限学习机构建MEMS陀螺误差补偿模型。通过实例研究，结果显示该方法能良好地补偿随机误差，与其他方法比较，具有更好的效果。

Application of improved wavelet thresholding method and extreme learning machine in MEMS gyro random errors compensation

MEMS gyro random errors is one of the main factors affecting its accuracy. Aim at the problem of MEMS gyro random errors, this paper proposes a compensation method based on improved threshold function in wavelet de-noising combining with the model of extreme learning machine algorithm. The wavelet threshold function is used to improve the de-noising effect in wavelet de-nosing, and the MEMS gyroscope error compensation model is constructed by extreme learning machine. The experiments show that the results obtained in the current paper compensate well for random errors and outperformed other algorithms.