Please use this identifier to cite or link to this item:
Title: ADRIFT : autonomous navigation
Authors: Sammut, Jordan
Keywords: Collisions at sea -- Prevention
Rule of the road at sea
Fuzzy logic
Vehicles -- Automatic control
Issue Date: 2018
Citation: Sammut, J. (2018). ADRIFT: autonomous navigation (Bachelor's dissertation).
Abstract: An estimated 75%-96% of marine accident are accounted to be directly linked to human negligence[green1, green2]. This figure, along with the steady increase of vessel collisions over the years, has caused a rise in interest in autonomous navigation in relation to marine life. This proposal aims to shed light on some of the current technologies being researched within the relevant field. The COLREGs are a set of maritime laws which must be appropriately obeyed in order to safely prevent collision at sea[green3]. The regulations are split into various categories, depending on the type of scenario between two vessels in a current space. In most scenarios, one of the vessel is designated as the Give Way vessel, while is the other is given the role of a Stand On vessel. The Give Way vessel must take appropriate action in order to steer clear of the Stand On one, which must maintain its current course, unless the Give Way vessel fails to take action. A system was proposed in order to showcase a vessel with autonomous capabilities when confronted with various collision scenarios. The system consisted of a simulator, in order to visualize the vessels' journeys. The outputs relating to the collision avoidance actions were calculated using a Fuzzy Inference System, which proved to be popular amongst recent literatures. Results relating to the system proved to be very promising, with a number of test cases relating to different scenarios, as showcased in the COLREGs rules and regulations, being evaluated. The successful functioning of the system justifies the use of the Fuzzy Logic approach utilized for the decision-making component of the software.
Appears in Collections:Dissertations - FacICT - 2018
Dissertations - FacICTAI - 2018

Files in This Item:
File Description SizeFormat 
  Restricted Access
1.39 MBAdobe PDFView/Open Request a copy

Items in OAR@UM are protected by copyright, with all rights reserved, unless otherwise indicated.