Implementação de um controlador com Ganho escalonado variantes no tempo em uma dinâmica ball-and-beam e em uma aeronave de asa móvel

Abstract

The object of study of this dissertation is a control technique that applies the gain scheduling strategy in a linear representation with time-varying parameters of a nonlinear system. During the controller synthesis, the dynamics of some state variables are approximated to a representation with time-varying parameters, by means of spatial discretization. The construction of the controller ensures stability, through convex conditions between the original and the approximate system, performing through state feedback. The method also has a mechanism to reduce the constraints of the search space, based on the adjacency relations between the vertices of the polytopes. Another contribution that this work presents is the development of new conditions, with less constraints in relation to the format of the matrices, thus enabling a reduction in complexity for numerical resolution of the computational algorithm. The study uses a ball and beam system for simulated experiments and a moving wing aircraft emulator as platform for practical tests.