陀螺阵列自适应加权数据融合算法  

Adaptive weighted data fusion algorithm for gyro array

作　　者：彭慧 黄俊[1,2] 黄江峰 刘宇[1,2] 林非凡 PENG Hui;HUANG Jun;HUANG Jiangfeng;LIU Yu;LIN Feifan(Chongqing Key Laboratory of Autonomous Navigation and Microsystems,Chongqing University of Posts and Telecommunications,Chongqing 400065,China;Chongqing Engineering Research Center of Intelligent Sensing Technology and Microsystem,Chongqing University of Posts and Telecommunications,Chongqing 400065,China)

机构地区：[1]重庆邮电大学自主导航与微系统重庆市重点实验室,重庆400065 [2]重庆邮电大学智能传感技术与微系统重庆市高校工程研究中心,重庆400065

出　　处：《中国惯性技术学报》2025年第3期213-218,228,共7页Journal of Chinese Inertial Technology

基　　金：国家自然科学基金(52175531);重庆市自然科学基金(CSTB2024NSCQ-LZX0126,CSTB2022NSCQ-LZX0050,cstc2022ycjh-bgzxm0190)。

摘　　要：微机电系统(MEMS)陀螺阵列信息融合可以提高测量精度。为提升融合数据精度,提出了一种基于自适应加权的冗余陀螺数据融合算法。首先采用箱线图算法识别并剔除传感器数据中的异常值,然后通过静态数据分析获取固定权值,同时通过改进Jousselme证据距离计算传感器置信距离,构建支持度函数,得到测量权值。其次,通过优化权重分配实现最优的数据融合。最后用8个MEMS陀螺组成的阵列对所提算法进行验证。实验结果表明:在静态下,陀螺三轴的角度随机游走平均改善约3倍、角速率随机游走平均改善约3.7倍、零偏不稳定性平均改善约3倍;在以1σ标准差为指标的动态实验中,所提方法的数据融合误差比均值滤波平均降低约28.5%,有效提高了陀螺阵列数据融合精度。The information fusion of microelectromechanical system(MEMS) gyro array can improve measurement accuracy.To enhance the accuracy of fused data,a redundant gyro data fusion algorithm based on adaptive weighting is proposed.Firstly,the box plot algorithm is used to identify and eliminate the outliers in the sensor data,and then the fixed weights are obtained through static data analysis.At the same time,the sensor confidence distance is calculated by improving the Jousselme evidence distance,and the support function is constructed to obtain the measurement weights.Secondly,the optimal data fusion is achieved by optimizing weight allocation.Finally,the proposed algorithm is validated by an array of 8 MEMS gyros.The experimental results show that the angular random walk,angular rate random walk,and zero-bias instability of the three axes of gyros are optimized by an average of about 3 times,3.7 times,and 3 times,respectively.Under dynamic experiments with 1σ standard deviation as the indicator,the data fusion error of the proposed method is reduced by about 28.5% on average compared to mean filtering,effectively improving the accuracy of gyro array data fusion.

关 键 词：冗余陀螺 支持度 自适应加权 权重分配 

分 类 号：U666.1[交通运输工程—船舶及航道工程]