基于EKF的多MEMS传感器姿态测量系统

姿态信息是飞行控制中最关键的参数之一,因此姿态测量成为飞行控制系统首要解决的问题。利用多M EM S传感器研制了一种微型姿态测量系统。利用三轴M EM S加速度计和三轴M EM S陀螺数据,由方向余弦矩阵的姿态表示形式推导了扩展K a lm an滤波方程,解算出飞行器的俯仰角和横滚角;设计专家系统判断飞行器的运动状态,并根据该状态调整滤波算法中的测量噪声矩阵,使系统可同时满足静态情况和动态情况的使用;利用空速和高度数据对俯仰角进行修正,利用GPS解算航向角。将实验结果与国外最新的商用自动驾驶仪的姿态结果进行了比较,二者在静态情况下非常吻合,在动态情况下基本吻合。

MEMS multi-sensor attitude determination system based on the EKF

The attitude information is one of the most important parameters in flight control systems.This paper describes an attitude measurement system based on a MEMS multi-sensor.The data from 3-axis MEMS accelerators and 3-axis MEMS gyros is used to determine the pitch angle and the roll angle of the air vehicle using an extended Kalman filter(EKF) equation deduced from the direction cosine matrix.An expert system was designed to estimate the vehicle motion to adjust the measurement noise matrix of the extended Kalman filter equation so that the system can be used in dynamic as well as static environments.The measured airspeed and altitude are also used to revise the pitch angle.The heading is determined from GPS information.Test results compare well with an imported commercial autopilot in a static test with less accurate results in a dynamic test.