Behavior based autonomous mobile Robot for industrial logistics

• Autor(a) principal: Corvo, João Nuno Salvador Rodrigues
• Data de Publicação: 2022
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10362/153181
• topic: autonomous mobile robots; behavior trees; industry 4.0; autonomous navigation; path planning; ROS; Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática

The design of robot behaviors to meet the requirements of the new industrial era - Industry 4.0 - has grown significantly in recent years. Especially the demand for flexible and adaptable systems has increased exponentially since intelligent robots started to be integrated into assembly lines and replace human activities. Tools such as Finite State Machines have proven to be an understandable and quick way to solve high-level problems in robotics; however, unmanageable when complexity rises. They become confusing and unreadable, making their modification and mainte- nance a problem. New tools, such as Behavior Trees, have emerged, creating modular, flexible, and adaptable systems without sacrificing readability with the increased com- plexity. The proposed architecture follows a hierarchical layered approach taking advantage of Behavior Trees, developing modular robot skills and system interfaces to create an autonomous behavior-based system. The software was implemented and tested in an Autonomous Mobile Robot capable of navigating complex environments and executing basic tasks. The results showed real advantages in using the layer-based approach, particularly giving the system modularity and increased flexibility capable of being easily improved and used in other systems. It was also concluded that Behavior Trees are an adequate tool for reactive systems in highly dynamic environments.