姿态航向参考系统中关联信息分级解耦方法  

Research on hierarchical decoupling of association in-formation inattitude heading reference system

作　　者：郑威 杨俊杰(指导) 陶坤宇 ZHENG Wei;YANG Junjie;TAO Kunyu(School of Electronic Information Engineering,Shanghai Dianji University,Shanghai 201306,China)

机构地区：[1]上海电机学院电子信息学院,上海201306

出　　处：《上海电机学院学报》2025年第1期45-51,共7页Journal of Shanghai Dianji University

摘　　要：为了解决无人机中惯性MEMS传感器普遍存在的零偏不稳定、数据耦合干扰等问题,提出了一种基于AHRS的扩展卡尔曼滤波算法的分级解耦方法。首先,基于多传感器融合滤波技术,采用分级滤波结构,一级滤波使用加速度计校正陀螺仪漂移,获得横滚角和俯仰角姿态估计;其次,二级滤波使用磁力计进行偏航角误差校正,同时屏蔽对俯仰角和横滚角的观测更新,综合得到载体的全姿态信息;最后,通过实验对其进行动静态测试,验证磁干扰下分级卡尔曼滤波器的抗干扰能力和姿态解算精度。实验结果表明:该方法实现了磁力计与加速度计观测分离,既能弥补陀螺仪长航时发散的问题,又能实现多重关联信息的解耦。In order to solve the problems of zero-bias instability and data coupling interference of inertial micro-electro-mechanical system(MEMS)sensors in unmanned aerial vehicles,an extended kalman filter(EKF)structure based on attitude and heading reference system(AHRS)is proposed.Based on the multi-sensor fusion technology,a hierarchical filtering structure is adopted.In the first-stage filtering,the accelerometer is used to correct the gyroscope drift,providing the roll and pitch angle attitude estimation.In the second-stage filter the magnetometer is used to correct yaw angle error,while excluding the observation updates of pitch and roll angle,thus obtaining the full attitude information of the carrier.Finally,dynamic and static experiments are conducted to evaluate the anti-interference capability and attitude calculation accuracy of the hierarchical kalman filter under magnetic disturbance.Experimental results demonstrate that the proposed method achieves observation separation between magnetometers and accelerometers,compensates for long-term gyroscope drift,and effectively decouples multi-source correlated information.

关 键 词：微机电系统 航姿参考系统 扩展卡尔曼滤波 组合导航 

分 类 号：TN713[电子电信—电路与系统]