Controle em ambientes interiores de veículos aéreos não tripulados

• Ano de defesa: 2017
• Autor(a) principal: Santos, Milton César Paes
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Espírito Santo; BR; Doutorado em Engenharia Elétrica; Centro Tecnológico; UFES; Programa de Pós-Graduação em Engenharia Elétrica
• Palavras-chave em Português: Veículos aéreos não tripulados (VANTs); Controle não linear; Fusão sensorial; Navegação autônoma; Robótica aérea; Controle automático; Robôs móveis; Engenharia Elétrica; 621.3
• Link de acesso: http://repositorio.ufes.br/handle/10/9690

This thesis proposes nonlinear controllers to be applied to unmanned aerial vehicles (UAVs) in positioning, trajectory-tracking and path-following missions in 3D airspace. In addition, a prototype of a platform conceived to allow using UAVs in indoor environments is shown, with several computational tools that allows using aerial robots in such classic navigation missions. In order to contextualize the developed tools, we highlight the development of algorithms for the detection and localization of objects in the environment and a decentralized sensorial fusion structure that is used to improve the measurement of the posture data (position and orientation) of the UAVs, in addition to detecting obstacles. Subsequently, the same fusion filter is used to combine PVTOL (Planar Vertical Takeoff and Landing) control signals, relaxing the aircraft motion constraints, previously limited to displacement along the Z axis and the XZ and YZ vertical planes. Complementarily, the implementation of navigation and control systems is based on two different dynamic models, one simple and the other detailed. The control systems are designed using classic model inversion techniques, which has shown to be an efficient methodology for the design of controllers for the developed applications, namely positioning, trajectory-tracking and path-following. Additionally, a collision avoidance controller, an adaptive controller, and a controller with maximum speed saturation are designed. The stability of such controllers is demonstrated by the asymptotic convergence of the control variables and by the fulfillment of the control objective during the simulations and experiments performed, as expected from the theoretical analysis.