Modelo distribuído bio-inspirado para planejamento de caminho e navegação de times de robôs
Ano de defesa: | 2021 |
---|---|
Autor(a) principal: | |
Orientador(a): | |
Banca de defesa: | |
Tipo de documento: | Dissertação |
Tipo de acesso: | Acesso aberto |
Idioma: | por |
Instituição de defesa: |
Universidade Federal de Uberlândia
Brasil Programa de Pós-graduação em Ciência da Computação |
Programa de Pós-Graduação: |
Não Informado pela instituição
|
Departamento: |
Não Informado pela instituição
|
País: |
Não Informado pela instituição
|
Palavras-chave em Português: | |
Link de acesso: | https://repositorio.ufu.br/handle/123456789/32782 http://doi.org/10.14393/ufu.di.2021.254 |
Resumo: | Cellular automata (AC) are bioinspired approaches that have recently been investigated for various applications, including robotics. An improved model based on CA rules and local decision flow is proposed and evaluated for distributed and decentralized navigation of teams of autonomous robots. The proposed model was called BioTeam: a bio-inspired distributed system for path planning and navigation of robot teams. In the first stage, the path planning is carried out by each robot independently. The goal is for each robot on the team to be able to build a short, collision-free path from its current position to its goal, trying to avoid unnecessary detours as much as possible. During displacement along the defined path, the decentralized control system present in each robot must be able to avoid any situation of possible collision and, above all, to overcome possible trajectory conflicts involving other robots, without jeopardizing its own efficiency and that of the other team members. Thus, each robot achieves its goal independently, without communication with other robots. Experiments were carried out to confirm the efficiency of the new techniques employed. Simulations using the Webots platform showed promising results in the navigation of teams with 10 to 50 robots. These results confirm that the Bioteam model is capable of planning smooth and short routes and in navigation the robots can overcome different conflict situations independently and decentrally in large and complex environments, with several robots moving at the same time. |