O emprego otimizado de sistemas de aeronaves remotamente pilotadas na Amazônia Azul com o uso da Inteligência Artificial

Abstract

This Course Conclusion Work deals about route optimization in the use of Remotely Piloted Aircraft Systems (RPAS) in maritime surveillance or vessel search, with the use of Artificial Intelligence (AI). The objective of this study is to contribute to the Navy's strategic Project called the Blue Amazon Management System (SisGAAz). The methodology adopted included a bibliographical and documentary research and interviews with professionals with relevant knowledge in their respective áreas of activity, in addition to the author's experience as a Naval Aviator. The theme is in line with the trinomial “Defence, Security and Development”, as it refers to the continuous improvement of the monitoring system of the Brazilian maritime space. The preservation of this heritage, in which there is a diversity of natural resources, is of major importance for Brazil, in order to ensure its exploration, including by future generations. The work begins with concepts and delimitations regarding maritime spaces and sovereign rights for the purpose of exploration, conservation and management of living or non-living natural resources of the waters, the seabed and its subsoil. Next, it discusses the Survey of the Brazilian Continental Shelf (LEPLAC) and the proposals submitted for its expansion, which increases the área of the Blue Amazon to 5.7 million km². Next, it deals with the importance of the country having a maritime área monitoring and protection system, including for carrying out Search and Rescue activities. Afterwards, it discusses the current situation of the use of RPAS within the scope of the Ministry of Defense, with emphasis on the operations of the Hórus Squadron, of the Brazilian Air Force, which has a system capable of flying at medium altitude and long range, with communication via link satellite data. Then, it explores the scientific study authored by Lima Filho et al (2022), whose subject, which is the focus of this paper, deals with the possibility of carrying out, from RPAS, a maritime search or surveillance of vessels suspected of illicit activities, such as illegal fishing, unauthorized research, drug and arms trafficking and environmental crimes, through the creation of an Artificial Neural Network (ANN), capable of classifying, predicting and recognizing vessel patterns. The conclusion indicates that the use of RPAS capable of, at least, medium altitude and long range, with satellite communication and with a Synthetic Aperture Radar (SAR), associated with AI, for surveillance of maritime areas of interest, search for suspect vessels or SAR search, prove yourself as a good option, because, when activated, such means can quickly reach the destination region. When found, they will be able to view, identify and record the images with their electro-optical sensor and transmit them, in real time, to the control station.