Navegação de robôs móveis ou veículos multi-articulados com realimentação de informações visuais

• Ano de defesa: 2008
• Autor(a) principal: Fernandes, Mariana Rampinelli
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Espírito Santo; BR; Mestrado em Engenharia Elétrica; Centro Tecnológico; UFES; Programa de Pós-Graduação em Engenharia Elétrica
• Palavras-chave em Português: Processamento de imagens - Técnicas digitais; Sistemas de veículos auto-guiados; Sistemas difusos; Navegação de robôs móveis; Robôs móveis; Visão por computador; Medidas Elétricas, Magnéticas e Eletrônicas; Instrumentação; 621.3
• Link de acesso: http://repositorio.ufes.br/handle/10/4063

This work deals with the use of computer vision as the feedback for the fuzzy control based in inference propagation. This approach is a part of a supervisory system development to automate navigation navigation and/or maneuver’s control of mobile robots or multi-articulated vehicles. Those robots have a mechanical chain with the first element as a tractor, with control of direction’s angle in the front wheels, and speed control in the back wheels. The other elements of the chain are passive traillers. Such vehicles can be used to transport load, for example. The problem of control these robots is particularly complex in the backward movements, where the system is non-holonomic, behaving as a horizontal multiple pendulum, nonlinear, unstable with complex modelling. The computer vision was chosen to feedback the fuzzy controller due to the amount of informations about the work enviroment and about the vehicles itself that the images of a single camera can provide. For the extraction of the robot characteristics, the strategy chosen was the color segmentation. The experiments used a multi-articulated vehicle prototype which one has a control system and a embedded sensing with radio frequency communication that provides a link with a remote computer. The camera, base of the vision system is placed on the top of the tests’ area and has a connection with the remote computer. From the experimental results, was possible to verify that the controller with visual feedback is capable to follow the references. It was also noted that the control system performed satisfactorily with the visual feedback to align the vehicle with any initial configuration, either at the center or at the edge of the camera’s field of vision.