Advanced search
Start date
Betweenand


Robust control of quadcopters by means of linear matrix inequalities

Full text
Author(s):
Marciano Prates Salbego
Total Authors: 1
Document type: Master's Dissertation
Press: Campinas, SP.
Institution: Universidade Estadual de Campinas (UNICAMP). Faculdade de Engenharia Elétrica e de Computação
Defense date:
Examining board members:
Ricardo Coração de Leão Fontoura de Oliveira; Bruno Augusto Angélico; Matheus Souza
Advisor: Ricardo Coração de Leão Fontoura de Oliveira
Abstract

The problem of attitude stabilization of a quadcopter is investigated in this dissertation. The kinematics of the mechanical system is described in terms of the Cayley-Rodrigues parametrization, from which is constructed a quasi-LPV system that is used in the synthesis of gain-scheduled controllers associated to performance criteria. A set of nonlinear differential equations describing the rigid body dynamics is obtained for the validation of the proposed control strategy using the Euler-Lagrange formalism, that considers the kinematics described by quaternions. As additional contribution, it is investigated the design of L_1 adaptive controllers, foreseeing future applications of the adaptive controllers in the quadcopter attitude control problem. The synthesis of the controllers and filters in both investigated approaches is formulated in terms of parameter-dependent linear matrix inequalities. As solution method, polynomial approximations (relaxations) of arbitrary degree are employed, resulting in convex optimization problems of increasing precision based on linear matrix inequalities, that in this dissertation are obtained with the aid of specialized computational toolboxes. The advantages of the proposed approaches are illustrated by means of numerical comparisons with other techniques available in the literature (AU)

FAPESP's process: 14/11593-5 - Robust control of quadrotors by means of linear matrix inequalities
Grantee:Marciano Prates Salbego
Support Opportunities: Scholarships in Brazil - Master