| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 6.11 MB | Adobe PDF |
Authors
Abstract(s)
Com o fim da Guerra fria, o interesse pela guerra submarina diminuiu drasticamente. Contudo, recentes desenvolvimentos em veículos não tripulados e vigilância acústica ativa e passiva trouxeram a guerra submarina de volta ao centro das atenções. A guerra anti-submarina é uma das principais preocupações de uma marinha uma vez que é difícil detetar um submarino no vasto oceano. Conjugando os recentes desenvolvimentos em veículos não tripulados e vigilância acústica ativa e passiva, podemos realizar uma fusão de dados e aumentar o nosso conhecimento sobre os eventos que ocorrem nas nossas águas.
Os dados processados provenientes de sistemas de vigilância acústica podem potencialmente constituir uma fonte essencial de informações navais. Uma barreira acústica pode efetuar a deteção e classificação de contactos com sucesso. No entanto, estes sistemas requerem pessoal altamente qualificado para operar, apresentam uma infraestrutura dispendiosa e são difíceis de implementar e manter. A fusão de dados de várias fontes, e mesmo de sensores de baixo custo com medidas ruidosas, é uma solução promissora, especialmente se a otimização de recursos for uma prioridade. Neste contexto, a presente dissertação pretende ser uma prova de conceito de uma implementação de baixo custo em águas pouco profundas que pode ser facilmente expandida e evoluída para diferentes cenários.
Os testes iniciais de campo e de processamento, que decorreram no exercício Robotics Exercise (REX) 2022 e Fibersense, coordenado pelo Centro de Investigação Naval (CINAV), apresentam resultados promissores, mas há muito a melhorar para aumentar as capacidades e aplicabilidade do sistema.
With the end of the Cold War, the interest in underwater warfare decreased dramatically. However, recent developments in unmanned vehicles and active and passive acoustic surveillance have brought underwater warfare back to center stage. Anti-submarine warfare is one of the major concerns of a navy since it is difficult to detect an enemy submarine in the vast ocean. Conjugating the recent developments in unmanned vehicles and active and passive acoustic surveillance, we can perform data fusion and increase our knowledge about the events occurring in our waters. The processed data originating from acoustic surveillance can potentially be an essential source of naval intelligence. An acoustic barrier can perform this detection with success. Still, these systems require highly qualified personnel to operate, present a costly infrastructure, and are hard to implement and maintain. Data fusion from multiple sources, and even from low-cost sensors with noisy measures, is a promising solution, especially if resource optimization is a priority. On this basis, this dissertation is intended to be a proof-of-concept of a low-cost shallow-water implementation that can be easily expanded and evolved for different scenarios. The initial field and processing tests, that took place in the Robotics Exercise 2022 (REX) and Fibersense exercises, coordinated by the Center for Naval Research (CINAV), show promising results, but there is much room for improvement to increase the system’s capabilities and applicability.
With the end of the Cold War, the interest in underwater warfare decreased dramatically. However, recent developments in unmanned vehicles and active and passive acoustic surveillance have brought underwater warfare back to center stage. Anti-submarine warfare is one of the major concerns of a navy since it is difficult to detect an enemy submarine in the vast ocean. Conjugating the recent developments in unmanned vehicles and active and passive acoustic surveillance, we can perform data fusion and increase our knowledge about the events occurring in our waters. The processed data originating from acoustic surveillance can potentially be an essential source of naval intelligence. An acoustic barrier can perform this detection with success. Still, these systems require highly qualified personnel to operate, present a costly infrastructure, and are hard to implement and maintain. Data fusion from multiple sources, and even from low-cost sensors with noisy measures, is a promising solution, especially if resource optimization is a priority. On this basis, this dissertation is intended to be a proof-of-concept of a low-cost shallow-water implementation that can be easily expanded and evolved for different scenarios. The initial field and processing tests, that took place in the Robotics Exercise 2022 (REX) and Fibersense exercises, coordinated by the Center for Naval Research (CINAV), show promising results, but there is much room for improvement to increase the system’s capabilities and applicability.
Description
Keywords
Vigilância Acústica Aplicações acústicas Classificação de sinais múltiplos Algoritmos de processamento de sinais