SCKF for MAV attitude estimation

In this paper, a novel quaternion-based attitude estimation algorithm with the square-root cubature Kalman filter (SCKF) is proposed to improve the attitude estimation performance for micro air vehicle (MAV). The SCKF is a kind of new effective method to solve nonlinear state estimation, it can directly deal with nonlinear systems, and the QR decomposition in SCKF avoids the square-root operation of the covariance matrix, which ensures the continuity of the filtering process. We use quaternion to represent the attitude, treat the strap down attitude as state update, and adopt the bi-vector method to update the measurement vector. Simulation example shows that the proposed algorithm can get better performance on the estimate accuracy and error robustness than that of extended Kalman filter (EKF) and unscented Kalman filter (UKF).