In-flight misalignment attitude estimation for UAV based on GPS and UKF/MPF filter

In view of the long alignment time of ground static base for unmanned aerial vehicle (UAV), an in-flight misalignment attitude estimation method for airborne SINS was presented which was based on GPS observations and a novel model predict filter (MPF). The estimation procedure can be divided into two stage, i.e., coarse estimation stage during the uniform straight-line motion of aircraft, in which GPS was used to obtain the initial position, velocity and azimuth of SINS; accurate estimation stage during a S flight of aircraft in which an MPF combined with the unscented Kalman filter (UKF) was used to estimate the error of SINS due to the coarse estimation. Simulation results demonstrate that the misalignment attitude estimation method is feasible.