Non-linear estimation of the flexural lever arm for transfer alignment of airborne distributed position and orientation system

As the development of aerial mapping, obtaining the motion parameters of the mapping sensors mounted under the flexural wing of the aircraft is necessary to accomplish the mission. Airborne distributed position and orientation system (DPOS) could be applicable which is composed of master inertial navigation system (INS), several slave INSs and the global positioning system. The slave INS is mounted on the mapping sensor for measuring the motion parameters. However, the flexural deformation of the wing will impose critical influence on the transfer alignment between master and slave INSs. The formed flexural lever arm should be compensated precisely. A non-linear relation between the flexural angle and flexural lever arm variation is proposed, and a novel 27-state non-linear model is established for transfer alignment of airborne DPOS. The nonlinear measurement equation of angular rate is derived based on three successive rotations of the body frame of the master INS. Unscented Kalman filter is utilised as the non-linear estimator for its capability of non-linear approximation. It is shown from the simulation results that the proposed method can estimate the flexural angles and flexural lever arm variations more accurately, and the attitude and position accuracy of slave INS is also improved correspondingly.