CAJ | 학술논문

针对小型四旋翼飞行器航姿测量系统中MEMS器件精度低、易发散等问题,提出一种新的数据融合方法,对航姿测量系统中陀螺仪、加速度计和电子罗盘的输出数据进行融合.利用MEMS惯性测量元件搭建了一个低成本航姿测量系统,分析了姿态解算的过程和难点,采用基于梯度下降的姿态融合算法,对四元数更新方程进行了补偿.实验结果表明:与目前常用的卡尔曼滤波算法相比,基于梯度下降的数据融合算法能显著降低对处理器速度和精度的要求,能有效融合航姿测量单元的传感器数据,提高小型四旋翼飞行器的姿态测量精度.
침대소형사선익비행기항자측량계통중MEMS기건정도저、역발산등문제,제출일충신적수거융합방법,대항자측량계통중타라의、가속도계화전자라반적수출수거진행융합.이용MEMS관성측량원건탑건료일개저성본항자측량계통,분석료자태해산적과정화난점,채용기우제도하강적자태융합산법,대사원수경신방정진행료보상.실험결과표명:여목전상용적잡이만려파산법상비,기우제도하강적수거융합산법능현저강저대처리기속도화정도적요구,능유효융합항자측량단원적전감기수거,제고소형사선익비행기적자태측량정도.
For overcoming low precision and easy to divergence of MEMS inertial measurement sensors of small size quad-rotor aircraft attitude measurement, a new data fusion method is designed to fuse the information from gyroscope, accelerometer and magnetometer. Use the MEMS inertial measurement components to construct a low-cost attitude heading measurement system, analyze the attitude estimation process and difficulties. Based on gradient descent, use attitude fusion algorithm to compensate quaternion updating equation. Experimental results show that the data fusion method based on the gradient descent algorithm is better than the Kalman filtering algorithm, and can significantly reduce the requirements of precision and speed for the processor, effectively fuse sensor data and improve the accuracy of attitude estimation of the small quad-rotor.