Eventos Anais de eventos
COBEM 2021
26th International Congress of Mechanical Engineering
LQR-Backstepping hybrid controller for flight stabilization for autonomous quadrirotor
Submission Author:
Douglas Arena , SP , Brazil
Co-Authors:
Douglas Arena, Magno Enrique Mendoza Meza, marcos fernandes, Elvira Rafikova
Presenter: Douglas Arena
doi://10.26678/ABCM.COBEM2021.COB2021-0543
Abstract
In the last few years, there has been an exponential growth in the use of unmanned aerial vehicles (UAVs) of the quadrotor type in several scenarios, due to its ability to perform different types of tasks with reduced risk to human life and good cost-benefit ratio. Quadrotors are being used in increasingly complex applications, which drove the emergence and development of platforms aiming to carry out tasks in real time with greater autonomy. The maneuverability of quadrotors faces great challenges related to the non-linear dynamics of quadrotors, as well as the underactuated and fast dynamic systems. In view of the conditions and difficulties imposed by the dynamics, it is necessary to choose a control technique that is capable of leading the quadrotor to follow a trajectory in a stable manner. The control law proposed in this work is the combination of LQR (Linear Quadratic Regulator) and backstepping control techniques. The first being a linear control technique based on optimal control and the second a non-linear technique based on Lyapunov's stability criteria. The proposed control law will be called LQR-Backstepping. In order to reduce control efforts, the LQR control technique will be used to obtain virtual controls, being responsible for determining the control angles, whose objective is the orientation of the translation movements, resulting in displacement to the desired positions in X, Y and Z. Also aiming to stabilize and guarantee the convergence to the desired trajectory, the backstepping control will be responsible for controlling the Euler angles, therefore, it will allow the convergence of the angles to the oriented positions (control angles). The simulation results showed that the LQR-Backstepping control shows satisfactory results in energy efficiency and trajectory tracking, although the quadrotor is a non-linear and underactuated system. Thus, we can conclude that the technique presented is appropriate and effective in terms of energy and traceability requirements. The proposed control law is not commonly found in the control literature for application in AUV (quadrotor), therefore, it is intended that the quadrotor flight control area be extended with the control law proposed in this work.
Keywords
Autonomous UAVs, LQR controller, Backstepping control, non linear control
