Planejamento de movimento para robôs móveis baseado em uma representação compacta da Rapidly-Exploring Random Tree (RRT)

• Ano de defesa: 2017
• Autor(a) principal: Sousa, Stephanie Kamarry Alves de
• Orientador(a): Molina, Lucas
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Sergipe
• Programa de Pós-Graduação: Pós-Graduação em Engenharia Elétrica
• País: Brasil
• Palavras-chave em Português: Engenharia elétrica; Robótica; Planejamento experimental; Algoritmos de computador; RRT; PRM; Planejamento de movimento; Planejamento probabilístico
• Área do conhecimento CNPq: ENGENHARIAS::ENGENHARIA ELETRICA
• Link de acesso: https://ri.ufs.br/handle/riufs/5034

The evolution of mobile robotics has directed research in this area to solve increasingly complex tasks. In these tasks, when optimized behaviors are specified, a deliberative process is required in order to determine the best action before executing it. In navigation architectures, the deliberation process is usually accomplished by a motion planning strategy. One of the motion planning techniques which has received much of the attention from the researches is the Rapidly-exploring Random Tree (RRT), because of its capacity to reduce representation dimension quickly. The vast majority of the research developed in this area, so far, is mainly focused on developing variants of the RRT for specific problems, not providing detailed analyzes regarding the influence of different variables in the classical algorithm. In this master’s work the focus is precisely to fill this gap by investigating the influence of different variables that compose the classic RRT algorithm, in other words, a detailed analysis of the RRT degrees of freedom and its influence on the final result. In addition, unlike most RRT papers, where the objective is to find the best path between two points, this dissertation presents a new approach in RRT searches by combining the search for a compact and complete representation of the configuration space with a low computational cost and knowledge of only the robot’s goal configuration. To validate and analyze the results obtained, tests by simulation are performed.