Controlador NMPC para seguimento de caminhos visuais aplicados a um VANT

Abstract

Nowadays, control strategies utilizing UAVs and cameras are used in several industrial applications and for example in agricultural, energy companies, hobbies, and many others. So indoors and outdoors environments are part of these applications. The main objective of this thesis is to propose a control strategy applied to UAVs using line-following and computer vision. In this way, this paper presents a Non-linear Model Predictive Control (NMPC) combined with a PID controller aiming to provide smooth control actions. To evaluate the model, the system was put into a circuit with colored segments and controlled by visual information utilizing the HSV colors. The authors perform simulations and real tests, in environments containing flaws in the visual segments. The control scheme presented in this work has a better performance when compared to the NMPC standalone or classic control strategies such as PID.