Maritime modular anomaly detection framework

• Autor(a) principal: Machado, Tomás Manuel Cardoso
• Data de Publicação: 2018
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10071/17590
• topic: Maritime framework; Anomaly detection; AIS data; Engenharia informática; Segurança marítima; Prevenção de riscos; Método de detecção; Análise de dados; Desenho de sistemas

Detecting maritime anomalies is an extremely important task for maritime agencies around the globe. With the number of vessels at seas growing exponentially, the need for novel automated methods to support them with their routines and upgrade existing technologies is undeniable. MARISA, the Maritime Integrated Surveillance Awareness project, aims at fostering collaboration between 22 governmental organisations and enhance the reaction and decision-making capabilities of the maritime authorities. This work describes our contributions to the development of MARISA’s common toolkit for the detection of maritime anomalies. These efforts, as part of a Masters’ dissertation, lead to the development of the Modular Anomaly Detection Framework, MAD-F, a full data pipe-line which applies efficient and reliable routines to raw vessel navigational data in order to output potential maritime vessel anomalies. The anomalies considered for this work were defined by the experts from various maritime institutions, through MARISA, and allowed us to implement solutions given the real needs in the industry. The MADF functionalities will be validated through actual real maritime exercises. In its current state, we believe that the MAD-F is able to support maritime agencies and be integrated into their legacy systems.