基于注意力机制与1D-CNN-Bi-GRU的MEMS陀螺仪降噪方法研究  

Noise Reduction Method for MEMS Gyroscope Based on Attention Mechanism and 1D-CNN-Bi-GRU

作　　者：王宁 武雷 强敏利 王凯 郭英超 WANG Ning;WU Lei;QIANG Minli

机构地区：[1]陕西华燕航空仪表有限公司,西安710100

出　　处：《科技创新与应用》2024年第26期37-41,共5页Technology Innovation and Application

摘　　要：为降低MEMS陀螺仪输出信号噪声,提出一种基于注意力机制和1D-CNN-Bi-GRU的MEMS陀螺仪误差补偿方法。首先通过1D-CNN网络进行陀螺时序数据特征提取,然后采用Bi-GRU网络进行特征分析,最后利用注意力机制进行数据权重分配。实验结果表明,与对照方法相比,该文方法噪声平均值减少57.03%和47.8%,噪声方差减少81.07%和73.01%,表明该文方法具有一定优势,具有较好的误差补偿效果。In order to reduce the output signal noise of MEMS gyroscope,an error compensation method of MEMS gyroscope based on attention mechanism and 1D-CNN-Bi-GRU is proposed.Firstly,the feature of gyro time series data is extracted by 1D-CNN network,then the feature is analyzed by Bi-GRU network,and finally the data weight is distributed by attention mechanism.The experimental results show that,compared with the control method,the average noise of this method is reduced by 57.03%and 47.8%,and the noise variance is reduced by 81.07%and 73.01%,indicating that this method has certain advantages and has a better error compensation effect.

关 键 词：MEMS陀螺仪 降噪 1D-CNN Bi-GRU 注意力机制 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]