Autonomous and Cooperative Pathfinding Technique for Swarms of Unmanned Aerial Vehicles in Dynamic Environments

• Ano de defesa: 2024
• Autor(a) principal: Tinoco, Claudiney Ramos
• Tipo de documento: Tese
• Tipo de acesso: Acesso embargado
• Idioma: eng
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Ciência da Computação
• Palavras-chave em Português: Swarm Robotics; Robótica de Enxames; Swarm Coordination; Coordenação de Enxames; Swarm Intelligence; Inteligência de Enxames; Emergent Behaviour; Comportamento Emergente; Pathfinding; Busca de Caminhos; Search-and-Rescue; Busca e Resgate; Dynamic Environments; Ambientes Dinâmicos; Drones; Unmanned Aerial Vehicles; Veículos Aéreos Não Tripulados; Wildfires; Incêndios Florestais; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO; Computação; ODS::ODS 11. Cidades e comunidades sustentáveis - Tornar as cidades e os assentamentos humanos inclusivos, seguros, resilientes e sustentáveis.; ODS::ODS 13. Ação contra a mudança global do clima - Tomar medidas urgentes para combater a mudança climática e seus impactos.; ODS::ODS 15. Vida terrestre - Proteger, recuperar e promover o uso sustentável dos ecossistemas terrestres, gerir de forma sustentável as florestas, combater a desertificação, deter e reverter a degradação da Terra e deter a perda da biodiversidade.
• Link de acesso: https://repositorio.ufu.br/handle/123456789/42218; http://doi.org/10.14393/ufu.te.2024.575

The coordination of swarms of robots is a task that demands the integration of various components, becoming even more complex in the case of swarms of Unmanned Aerial Vehicles (UAVs). This process involves a range of challenges, such as managing the flight controllers of each UAV, establishing effective intra-swarm communication mechanisms, ensuring adequate coordination and cooperation, and handling subtasks like pathfinding. In this context, this work proposes a pathfinding technique specifically adapted for swarm robotics. Since path construction is carried out incrementally and locally, i.e., the robots cooperate with each other and the path is found through the emergence of the swarm's global behaviour, the proposed technique becomes fully adaptable not only to static environments but also to dynamic ones. This adaptability is essential in dynamic scenarios, where conditions can change rapidly, requiring a swift and coordinated response from the swarm. The developed technique is validated through a coordination model for swarms of UAVs, also proposed in this work, with the aim of assisting in the evacuation of individuals from wildfire-affected areas. In high-risk situations like this, speed and efficiency in identifying safe escape routes can be crucial for saving lives. In this regard, the UAVs, by working collaboratively, not only identify but also converge on viable escape paths, effectively signalling them to individuals in danger. The outcomes obtained with the proposed technique and coordination model are promising. Experiments conducted in different types of environments, as well as with swarms of various sizes, demonstrated the emergence of a robust global behaviour capable of executing both the search and delineation of safe paths, thereby facilitating the evacuation of individuals from hazardous areas. This advance represents a significant contribution to swarm robotics applied to the safety of individuals, paving the way for future innovations in this context.