基于UKF的MEMS传感器姿态测量系统  

MEMS Sensors Based Attitude Measurement System Using UKF

作　　者：赵鹤[1] 王喆垚[1] 

机构地区：[1]清华大学微电子学研究所,北京100084

出　　处：《传感技术学报》2011年第5期642-646,共5页Chinese Journal of Sensors and Actuators

基　　金：ABB(中国)有限公司研究中心项目

摘　　要：针对工业和民用领域对姿态测量的需求,提出了基于MEMS加速度计、陀螺仪和磁强计的姿态测量系统,并采用无先导卡尔曼滤波(Unscented Kalm an F ilter,UKF)方法处理传感器数据。针对基于加速度计和磁强计的姿态测量方式在动态测量时不准确的问题和单独采用陀螺仪测量角度产生漂移的问题,设计了基于方向旋转矩阵的UKF滤波器:结合两种测量方式的优点,互相克服各自的不足。经过UKF处理后:在静态测试中,测量系统没有积累误差;在动态测试中,陀螺仪补偿了加速度计和磁强计测量方法的中产生的毛刺。实验结果表明,所设计的姿态系统可以达到很高的精度;通过嵌入式系统完成数据采集和滤波,姿态测量系统具有很好的实时性。To satisfy the requirements of industry and civil use,attitude measurement system,based on MEMS accelerometer,gyroscope and magnetometer,is proposed.Unscented Kalman Filter（UKF）is used to fuse the data.To solve the drift of gyroscope and inaccuracy in dynamic using accelerometer and magnetometer unit,the UKF,based on rotation matrix,is addressed to combine the advantages of the two measurement methods and to overcome the faults of each other＇s.The accumulated errors are calibrated and the accuracy is improved in static state,and the faults in dynamic state are compensated.The experimental results show that the attitude system can achieve high accuracy.Since the data acquisition and filtering are implemented by the embedded system,our system can work in real time.

关 键 词：惯性导航 姿态测量 微机电系统(MEMS) 无先导卡尔曼滤波(UKF) 

分 类 号：TP212.9[自动化与计算机技术—检测技术与自动化装置] V249[自动化与计算机技术—控制科学与工程]