UAV Navigation System for Prescribed Fires

• Autor(a) principal: Muchaxo, Nuno Miguel Pereira
• Data de Publicação: 2020
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10362/118785
• topic: Forest Fire; Unmanned Aerial Vehicle; Navigation; Planner; Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática

Since the beginning of mankind, a lot of fires have happened and have taken millions of lives, whether they were human or animal lives. On average, there are about twenty thousand forest fires annually in the world and the burnt area is one per thousand of the total forest area on Earth. In the last years, there were a lot of big fires such as the fires in Pedrogão Grande, Portugal, the SoCal fires in the US coast, the big fire in the Amazon Forest in Brazil and the bush fires in Australia, later 2019. When fires take such dimensions, they can also cause several environmental and health problems. These problems can be damage to millions of hectares of forest resources, the evacuation of thousands of people, burning of homes and devastation of infrastructures. When a big fire starts, the priority is the rapid rescue of lives and then, the attempt to control the fire. In these scenarios, autonomous robots are a very good assistance because they can help in the rescue missions and monitoring the fire. These autonomous robots include the unmanned aerial vehicle, or commonly called the UAV. This dissertation begins with an intensive research on the work that has already been done relative to this subject. It will then continue with the testing of different simulators and see which better fits for this type of work. With this, it will be implemented a simulation that can represent fires and has physics for test purposes, in order to test without causing any material damage in the real world. After the simulation part is done, algorithm testing and bench marking are expected, in order to compare different algorithms and see which are the best for this type of applications. If everything goes according to plan, in the end, it is expected to have an autonomous navigation system for UAVs to navigate through burnt areas and wildfires to monitor the development of these.