基于改进自适应滤波的MEMS陀螺振动误差抑制研究  

Research on vibration error suppression of MEMS gyroscope based on improved adaptive filtering

作　　者：陈杰[1,2] 侯帅康 刘玉县 何春华 CHEN Jie;HOU Shuaikang;LIU Yuxian;HE Chunhua(School of Electrical and Information Engineering,Zhengzhou University,Zhengzhou 450001,China;Guangdong Shunde Industrial Design and Research Institute(Guangdong Shunde Innovation Design and Research Institute),Foshan 528311,China;School of Integrated Circuits,Peking University,Beijing 100871,China)

机构地区：[1]郑州大学电气与信息工程学院,河南郑州450001 [2]广东顺德工业设计研究院(广东顺德创新设计研究院),广东佛山528311 [3]北京大学集成电路学院,北京100871

出　　处：《传感器与微系统》2024年第4期61-63,67,共4页Transducer and Microsystem Technologies

基　　金：国家自然科学基金资助项目(62104047)。

摘　　要：本文提出一种基于改进自适应滤波的MEMS陀螺随机振动误差的补偿算法。该算法采用简化Sage-Husa自适应滤波算法估计量测噪声,并通过协方差匹配技术引入收敛性判据抑制了滤波的发散,它提高了实时性且减少了计算量。实验结果表明:经过改进算法滤波后,MEMS陀螺随机振动误差的方差减少97.76%,与常规卡尔曼滤波相比,改进算法的方差减少了72.66%,验证了改进的自适应卡尔曼滤波算法可以有效地抑制MEMS陀螺因随机振动引起的输出误差。A compensation algorithm for random vibration error of MEMS gyroscope based on improved adaptive filtering is proposed.The algorithm uses a simplified Sage-Husa adaptive filtering algorithm to estimate measurement noise and convergence criterion is introduced to suppress filtering divergence by the covariance matching technique,which improves the real-time performance and reduces the computation quantity.The experimental results show that the variance of random vibration error of MEMS gyroscope is reduced by 97.76%after the algorithm filtering is improved,and the variance of the improved algorithm is reduced by 72.66%,compared with the conventional Kalman filtering.It is verified that the improved adaptive Kalman filtering algorithm can effectively suppress the output error caused by the random vibration of the MEMS gyroscope.

关 键 词：MEMS陀螺 简化Sage-Husa自适应滤波 时间序列模型 

分 类 号：TN98[电子电信—信息与通信工程] TP212[自动化与计算机技术—检测技术与自动化装置]