Aircraft attitude estimation based on central difference Kalman filter

When the extended Kalman filter (EKF) is applied in the aircraft attitude estimation, two defects exist: one is computational complexity; the other is large linearization error. Aiming at these problems, central difference Kalman filter (CDKF) based on Stirling interpolation formulation is applied to the low-cost aircraft attitude estimation system which is of less accurate and high noisy sensors. First, the nonlinear mathematical model of aircraft attitude based on quaternion is established, then CDKF is applied to attitude estimation. Experimental results with real flying data show that CDKF is superior to the commonly used EKF method and unscented Kalman filter (UKF). The algorithm not only improves the attitude estimation precision and stability effectively, but also avoids the computing burden of Jacobian matrices. In addition, it is more simple and easy to implement, because it has only one adjustable parameter instead of three in the UKF circumstances.