Αποφυγή σύγκρουσης αυτόνομου πλοίου με χρήση τεχνητού ανοσοποιητικού συστήματος

Παπαδόπουλος, Γιώργος; Papadopoulos, Georgios

dc.contributor.author	Παπαδόπουλος, Γιώργος	el
dc.contributor.author	Papadopoulos, Georgios	en
dc.date.accessioned	2022-11-30T11:33:19Z
dc.date.available	2022-11-30T11:33:19Z
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/56320
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.24018
dc.rights	Default License
dc.subject	Τομέας	el
dc.subject	Τεχνητό	el
dc.subject	Ανοσοποιητικό	el
dc.subject	αποφυγή	el
dc.subject	Ρίσκο	el
dc.subject	COLREG	en
dc.subject	Artificial	en
dc.subject	Immune	en
dc.subject	Risk	en
dc.subject	Collision	en
dc.title	Αποφυγή σύγκρουσης αυτόνομου πλοίου με χρήση τεχνητού ανοσοποιητικού συστήματος	el
heal.type	bachelorThesis
heal.classification	Collision Avoidance	en
heal.language	el
heal.access	free
heal.recordProvider	ntua	el
heal.publicationDate	2022-07-22
heal.abstract	The continuous development of technologies has pushed the shipping industry to develop automation and decision support systems, in order to handle and continuously monitor ships (AIS, GPS), in real time from any point on the planet, reducing much of the risk. Collisions are one of the most common types of maritime accidents, entailing significant risks to human life, the environment and economic benefit. The development of systems to avoid ship collisions seems to be able to help significantly in reducing a significant number of maritime accidents.In this thesis, a hypothetically autonomous collision avoidance algorithm based on the methodological framework of Artificial Immune Systems was developed. The primary goal was to develop such an algorithm and then to evaluate it through possible conflict scenarios, both in terms of security, as well as in terms of its execution time and finding the most economical path.The algorithm created uses hypothetical antibodies, which are generated after a risk is identified. As an antigen, it is considered a “generalized” cost function that we are required to minimize in order to eliminate risk.In the application, in the first frame, the planning of the ship's course on a given map is considered, considering the static obstacles of the area. The execution of this mission is implemented using the Velocity Obstacle algorithm, an algorithm that searches for the optimal path through motion mechanics.In thenext stage, the preservation of the path drawn by the algorithm is examined, detecting the risk of collision throughout the movement. In case a risk of collision is detected, the immune system algorithm (CLONALG) is activated, which is solely responsible for avoiding the collision. CLONALG considers an antigen a “generalized” cost function that it is asked to be minimizedin order to produce optimal antibodies, which are reflected as collision avoidance angles. Four different types of approach described inthe COLREGs were tested for completeness of application.The immune system algorithm seems to give satisfactory results for collision avoidance and a distance greater than the minimum to be maintained. The time to run and find the optimal value is closeto 6 seconds for each waypoint. Although the algorithm is not designed to conform to the COLREGs, it eventually conforms to them in some approximation situations. (el)	en
heal.advisorName	Ventikos, Nikolaos	en
heal.advisorName	Βεντίκος, Νικόλαος	el
heal.committeeMemberName	Ηλιοπούλου, Ελευθερία	el
heal.committeeMemberName	Γκίνης, Αλέξανδρος	el
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Ναυπηγών Μηχανολόγων Μηχανικών. Τομέας Μελέτης Πλοίου και Θαλάσσιων Μεταφορών	el
heal.academicPublisherID	ntua
heal.numberOfPages	92 σ.	el
heal.fullTextAvailability	false
heal.fullTextAvailability	false