Path Planning with Hybrid Maps for processing and memory usage optimisation

• Tipo: Artigo em Livro de Atas de Conferência Internacional
• Data: 2022

Autores

Luís C. Santos (Autor) Outra; Filipe Neves Santos (Autor) FEUP; André S. Aguiar (Autor) Outra; António Valente (Autor) Outra; Pedro Gomes da Costa (Autor) FEUP

Ata de Conferência Internacional

Título: IEEE International Conference on Autonomous Robot Systems and Competitions, ICARSC 2022, Santa Maria da Feira, Portugal, April 29-30, 2022

Páginas: 27-33

Indexação

ISI Web of Knowledge - 4 Citações

Scopus - 4 Citações

Outras Informações

ID Authenticus: P-00W-KW4

DOI: 10.1109/icarsc55462.2022.9784767

Abstract (EN): Robotics will play an essential role in agriculture. Deploying agricultural robots on the farm is still a challenging task due to the terrain's irregularity and size. Optimal path planning solutions may fail in larger terrains due to memory requirements as the search space increases. This work presents a novel open-source solution called AgRob Topologic Path Planner, which is capable of performing path planning operations using a hybrid map with topological and metric representations. A local A* algorithm pre-plans and saves local paths in local metric maps, saving them into the topological structure. Then, a graph-based A* performs a global search in the topological map, using the saved local paths to provide the full trajectory. Our results demonstrate that this solution could handle large maps (5 hectares) using just 0.002 % of the search space required by a previous solution.

Idioma: Inglês

Tipo (Avaliação Docente): Científica

Nº de páginas: 7