Desvio de obstáculos para UAVs em ambientes Dinâmicos

• Ano de defesa: 2024
• Autor(a) principal: Kim, Pedro Henrique Corrêa
• Orientador(a): Vivaldini, Kelen Cristiane Teixeira
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus Sorocaba
• Programa de Pós-Graduação: Programa de Pós-Graduação em Ciência da Computação - PPGCC-So
• Palavras-chave em Português: UAV; Drone; Robôs autônomos; Evitar colisões; Detecção de obstáculos
• Palavras-chave em Inglês: UAV; Drone; Collision avoidance; Autonomous Vehicles; Obstacle detection; Dynamic obstacles
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::METODOLOGIA E TECNICAS DA COMPUTACAO
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/19728

The growing use and accessibility of Unmanned Aerial Vehicles (UAVs) have provided an opportunity for automation in various applications. One area of great interest is the development of autonomous UAVs to perform tasks such as inspecting and exploring inaccessible or dangerous environments. Regardless of the application, the ability of UAVs to plan their own trajectory under different conditions is crucial. In this context, there is a significant effort to develop algorithms that ensure the autonomy of UAVs, allowing diverse interactions with the environment without the need for human intervention. Although there are various algorithms capable of performing trajectory planning in unknown and unstructured environments, the presence of dynamic obstacles poses a substantial challenge. The latency of sensors and the nature of existing algorithms limit their effectiveness in environments with moving obstacles. Given this scenario, this master's dissertation proposes the development of an innovative algorithm capable of addressing these challenges, using adaptive techniques from faster obstacle avoidance and obstacle detection algorithms. The goal is to provide a robust and reliable solution for autonomous trajectory planning in dynamic environments, where conventional methods face difficulties and may pose safety risks.