LPV and Nonlinear-based control of an Autonomous Quadcopter under variations of mass and moment of inertia

Abstract

This paper presents a hybrid robust control strategy to solve the trajectory tracking function of a quadcopter with time-varying mass. In the context of this paper, two ways of changing the mass of a quadcopter are considered. The first one involves a gradual continuous reduction of the mass throughout the flying time, and the second involves an abrupt change at some point during the flight. Besides the change of the mass, the moments of inertia with respect to the three axes are also changing. These moments of inertia are recalculated in realtime according to the mass changes.The quadcopter model is separated into two subsystems: rotational and translational. The rotational subsystem contains several time-varying parameters, such as the mass, the moments of inertia, and the speed of the rotor. It can be considered as a quasi-Linear Parameter Varying (LPV) system. An LPV H∞ controller has been designed to stabilize the orientation actuator’s dynamic. To ensure that the quadcopter follows the pre-defined trajectory, a combination of Integral Backstepping and Proportional Derivative (PD) controllers are used for the translational subsystem.The efficiency and robustness of the proposed cascaded controller with disturbances, noises, and model parameters uncertainties have been tested in MATLAB.