Autonomous aircraft maneuvering under unknown CG variations through a robust adaptive backstepping control

Abstract

The objective of the present article is to address the hitherto unaddressed problem of executing demanding autonomous aircraft maneuvers mitigating significant uncertainties in lateral center of gravity (c.g.) position. As a major contribution, it is shown that, under a reasonable simplifying assumption, a strict feedback form of the dynamics with affine in c.g. position can be derived from the highly coupled equations of motion that is caused by asymmetric c.g. variations. Thereafter, a two-step adaptive backstepping control is proposed adapting to the c.g. position automatically. Also, the useful nonlinearities in the dynamics are identified from the aircraft’s dataset and are retained to prevent a conservative control design. To provide further robustness to the aforementioned model simplification as well as to aerodynamic uncertainties, a fast adaptive sliding mode control is integrated with each of the two steps of the baseline adaptive backstepping control and the asymptotic stability of the overall system is proved using Lyapunov’s method. To validate the efficacy of the proposed control, the high angle of attack Herbst maneuver is simulated for the F18-HARV aircraft and nearly identical maneuver performance is achieved over a wide range of lateral c.g. variations on either side of the fuselage centerline. Real time hardware in loop simulations are also performed to establish the real time applicability of the proposed control.