Robust Adaptive Control Laws for Tilt-Rotor Quadcopters Subject to User-Defined Constraints

Abstract

This thesis focuses on Model Reference Adaptive Control (MRAC) and its application to a tilt-rotor quadcopter. After formulating two standard MRAC approaches, this thesis proposes a robust model reference adaptive control law that guarantees satisfactory trajectory following for the nonlinear dynamical system despite parametric, matched, and unmatched uncertainties. This control law is unique for its ability to exploit barrier Lyapunov functions and guarantee user-defined constraints both on the trajectory tracking error and the adaptive gains at all time. The proposed robust control law is then applied to design a control law for a tilt-rotor quadcopter with H-configuration with unknown and unsteady center of mass and matrix of inertia due to the presence of poorly modeled and dangling payloads. The tilt-rotor quadcopter equations of motion are presented and thoroughly analyzed. A novel approach is proposed to model the coupling between the translational and rotational dynamics as matched uncertainties, and a control strategy is developed to overcome the vehicle's underactuation. A tilt-rotor is designed and all of the components are presented and discussed. A challenging experiment where a tilt-rotor quadcopter pulls an unknown cart is performed and the results show the applicability of the proposed theoretical framework.