Automatic Route Planning and Calculation based on S-100 for Autonomous Vessels Navigation
Karamanoglou, Georgios (2022)
2022
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2022052712487
https://urn.fi/URN:NBN:fi:amk-2022052712487
Tiivistelmä
Automatic Route Planning and Distance Calculation can be further developed by using the same factors as used in Route Planning by a Navigator while using the new Hydrographic S-100 Standards. These can be used to make this process automatically while considering the weather, depth, navigational dangers, and other related information like a regular Navigator does in reality. The purpose of this Thesis is to explain how Route Planning and Distance Calculation works, show the implementation of the new S-100 IHO standard and most commonly used pathfinding algorithms in automatic route calculation. Finally, explain why accurate and adaptable route calculation is essential to Autonomous Ships and Shipping in general in the future.
For collecting the information for this Thesis, the chosen research methods were partly theoretical and partly constructive research, in the form of ascertaining and defining the problem to be solved, after presented the challenges that need to be tackled, literature review of researchers in the field of Pathfinding Algorithms applications, comparison of the literature collected, design of the process to solve the issue at hand, and theoretical evaluation of the various methods available depending on their use suitability and efficiency.
In conclusion, the technology is here for using the new S-100 IHO ENC Standard for Automatic Route Calculation while using, as the research has revealed out of a diverse set of pathfinding algorithms, ranging from Dijkstra to A*, genetic, including ant colony algorithms, which have been proven by empirical studies that they can be potentially employed in route navigation for ships. The adoption of Automatic Route Planning and Calculation in Autonomous Ship Navigation is advocated by using deep learning algorithms; further cybersecurity concerns are also present, including but not limited to signal jamming and malicious attacks on the ship's communication equipment. Nevertheless, diverse solutions have been also identified to mitigate the threats and ensure that the ships can successfully attain their goals in reaching targeted destinations. In particular, the use of neural networks and deep reinforcement learning algorithms is a prevalent solution.
For collecting the information for this Thesis, the chosen research methods were partly theoretical and partly constructive research, in the form of ascertaining and defining the problem to be solved, after presented the challenges that need to be tackled, literature review of researchers in the field of Pathfinding Algorithms applications, comparison of the literature collected, design of the process to solve the issue at hand, and theoretical evaluation of the various methods available depending on their use suitability and efficiency.
In conclusion, the technology is here for using the new S-100 IHO ENC Standard for Automatic Route Calculation while using, as the research has revealed out of a diverse set of pathfinding algorithms, ranging from Dijkstra to A*, genetic, including ant colony algorithms, which have been proven by empirical studies that they can be potentially employed in route navigation for ships. The adoption of Automatic Route Planning and Calculation in Autonomous Ship Navigation is advocated by using deep learning algorithms; further cybersecurity concerns are also present, including but not limited to signal jamming and malicious attacks on the ship's communication equipment. Nevertheless, diverse solutions have been also identified to mitigate the threats and ensure that the ships can successfully attain their goals in reaching targeted destinations. In particular, the use of neural networks and deep reinforcement learning algorithms is a prevalent solution.