Plataforma para experimentação de sistemas com múltiplos robôs em ambientes industriais

• Ano de defesa: 2019
• Autor(a) principal: Limeira, Marcelo Aurélio
• Tipo de documento: Dissertação
• 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: Robôs móveis - Controle automático; Inteligência coletiva; Internet das coisas; Inteligência artificial; Pesquisa operacional - Processamento de dados; Realidade aumentada; Arena (Programa de computador); Robótica; Mobile robots - Automatic control; Swarm intelligence; Internet of things; Artificial intelligence; Operations research - Data processing; Augmented reality; Arena (Computer file); Robotics; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA; Engenharia Elétrica
• Link de acesso: http://repositorio.utfpr.edu.br/jspui/handle/1/5102

Multi-Robot Systems (MRS) have been applied to the industrial sector extremely effectively, especially in intelligent warehouses, increasing efficiency and reducing operating costs. Most of the projects developed in this area are validated only in virtual simulators, making the implementation of these systems in real applications considerably difficult, due to their inability to faithfully reproduce all variables present in the real world. On the other hand, developing a totally real structure that meets the necessary conditions for the development of these systems could be both technically and financially unfeasible. This work presents an environment for experimentation of advanced behaviors in intelligent warehouses, allowing the experimentation of systems with multiple robots (MRS), interconnected, cooperative and interacting with virtual elements. The ARENA concept introduces a new approach to realistic and immersive experimentation in industrial environments, with the aim of evaluating new technologies aligned with Industry 4.0. The proposed method consists of reproducing a small-scale warehouse, inspired by a real scenario, managed by a group of autonomous, interconnected forklifts, incorporated by a ROS (Robot Operating System) architecture-compliant micro-robot swarm. Augmented Reality (AR) is employed to enhance the capabilities of these robots, allowing them to exhibit characteristics unfeasible by their size, such as the movement of goods and the adoption of certain sensors. Another application of RA is the insertion of resources that help in the accomplishment of the experiments, such as: marking of the traveled path, division of areas by color and marking of specific positions. Virtual Laser Range Finders (LRF) are specially designed from information from a global RGB-D camera, positioned over the environment, to improve robot perception, to detect obstacles and map the environment. All these features give the solutions developed through this platform a level of development very close to the level required for effective practical application.