Método colaborativo para posicionamento de precisão usando VANT e VTNT para a inspeção detalhada de torres de distribuição de energia elétrica

• Ano de defesa: 2020
• Autor(a) principal: Cantieri, Alvaro Rogerio
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Tecnológica Federal do Paraná; Curitiba; Brasil; Programa de Pós-Graduação em Engenharia Elétrica e Informática Industrial; UTFPR
• Palavras-chave em Português: Energia elétrica - Distribuição - Manutenção e reparos; Energia elétrica - Distribuição - Inspeção; Veículos autônomos; Sistemas de controle inteligente; Processamento de imagens; Visão por computador; Sensoriamento remoto; Sensores - Avaliação; Electric power distribution - Maintenance and repair; Electric power distribution - Inspection; Automated vehicles; Intelligent control systems; Image processing; Computer vision; Sensors - Evaluation; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO; Engenharia Elétrica
• Link de acesso: http://repositorio.utfpr.edu.br/jspui/handle/1/25104

The inspection process for the structural conditions of energy distribution pylons is a mandatory task defined by the National Electric Energy Agency – ANEEL in Brazil. A significant increase in the application of autonomous vehicles to inspect these structures is observed recently, including small unmanned aerial vehicles (UAVs), due to the operational advantages presented by these vehicles. The development of autonomous inspection architectures for this kind of application presents several technical challenges for the safe and effective performance of tasks related to the operation, including the obstacle detection and the precise positioning of vehicles during their displacement. The main objective of this work is to propose a cooperative method of positioning between an unmanned aerial vehicle (UAV) and an unmanned ground vehicle (UGV), based on a visual tag attached to the ground vehicle, for application in a detailed inspection of electricity distribution poles and pylons. This proposal sought to develop a method to increase the reliability of the autonomous displacement and route tracking from the vehicles during the inspection process, through the exchange of relative positioning information between them. Experiments performed to validate the UAV flight positioning data using the ground tag reference presents horizontal and vertical positioning errors lower than 2.0 meters, considering a confidence interval of 5 % for the measurements. The experiments executed to evaluate the navigation of the ground vehicle using the position data generated by the images captured by the UAV in flight, presented a positioning error lower than 0.25 m for all rounds, indicating the feasibility of using the cooperative navigation method proposed in real application conditions.