Planejamento de trajetórias para quadricópteros em tarefas de perseguição

• Ano de defesa: 2019
• Autor(a) principal: Lima, Gabriela Vieira
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Engenharia Elétrica
• Palavras-chave em Português: Quadricóptero; Quadrotor; Perseguição; Pursuit; Campos potenciais artificiais; Artificial potential field; Sistema de controle; Control system; Crazyflie 2.0; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA
• Link de acesso: https://repositorio.ufu.br/handle/123456789/27532; http://doi.org/10.14393/ufu.te.2019.2536

In order to accomplish increasingly complex tasks with higher performance, the autonomous and multi-agent systems have aroused the interest of researchers of the area of mobile robotics. In this context, this project develops an autonomous system, in such a way that an unmanned aircraft or a swarm can perform pursuit operations in a three-dimensional environment. Potential applications include surveillance and tracking in urban and natural environments. The supervisory system, implemented at a ground station, is responsible for planning viable trajectories for the pursuit process. The algorithm employed corresponds to the potential fields, and the designed route must ensure that there are no collisions with the obstacles in the workspace or between the agents. In order to enable trajectory tracking, each aircraft has a cascade control structure implemented in the firmware and a radio communication system. The aircrafts used in the experiments corresponds to Crazyflie 2.0 nano quadrotor, which due to its small size, is ideal for flights close to humans and indoor operations. This quadrotor has an inertial measurement unit (IMU), which combined with a local positioning system, makes it possible to estimate its position and orientation in three-dimensional space. Based on the strategies adopted, it was possible to obtain solutions to pursuit problems in different scenarios, considering convex and limited environments, cooperative operations and analysis in static and dynamic environments. For the experimental tests, the environment and its obstacles were previously mapped, as the aircraft do not have online mapping sensors.