Autonomous collision avoidance system for unmanned vessels: algorithms and software.

Abstract


Ship collision is one of the most substantial concerns in the global maritime transportation industry. Hence, navigation safety is considerably cited topic in maritime literature. Recently, Unmanned Navigation (UN) technology is gradually becoming more widely used across in the field of marine robotics. The paper investigates the problem of navigation safety in the movement control of Unmanned Vessels (UVs). The object of the study is the process of avoiding collisions of UVs. The subject of the research is the algorithms of the autonomous decision-making system and software for preventing vessel collisions during UN. The intent of this article is to improve the safety of UN by developing new Decision-Making algorithms for autonomous collision prevention of UVs in real time, taking into account the International Rules for the Prevention of Collisions at Sea, 1972 (COLREGs-72) and the recommendations of the Federal agency for sea and inland water transport of the Russian Federation (Rosmorrechflot).In this article, the fundamental concept and the key functions set of the Autonomous Collision Avoidance System (ACAS) are carried out for UVs which are marine transport vehicles capable of sensing its environment and operating without human involvement. Along this line of research, this work focuses on the development of a software algorithm for determining the most dangerous obstacle located within a radius of 12 miles (recommendations of Rosmorrechflot) around an UV based on the principle of vessels collision avoidance geometry, collision risk assessment and the characteristics of obstacles. Moreover, the proposed algorithms can prevent the collision and address the issues of real-time collision avoidance for UVs. The simulation results also demonstrate the promising application of the proposed algorithms in studying the UN safety. Nonetheless, this study provides a way forward to conduct a new information decision-making system design for UVs collision avoidance. This is currently under development, and will be proposed later.

Full Text

Improving the marine navigation and safety of sea transportation are a complex multi-level tasks, which are provided by the work of the International Maritime Organization (IMO), ship classification societies and administrations of governments involved in the development of world navigation. Analyzing vessel accidents, it can be noted that ship collisions, in particular, are one of the most common types of sea traffic accidents. Collision of vessels is the name given to an incident that occurred as a result of mutual contact between a vessel and another object during its movement and entailed loss of life or injuries, damage to the vessels and cargo, environmental pollution or other material damage.According to official data from the Federal service for supervision of transport of the Rus-sian Federation (Rostransnadzor), in the period from 2014 to 2020 collisions of vessels flying the State flag of the Russian Federation represent respectively 4.5 %, 1.4 %, 3.66 %, 7.14 %, 8.74 %, 3 % and 11.67 % of the total the number of accidents at sea. And also, were equal to 6.25 %, 11.83 %, 8.6 %, 8 %, 9.5 %, 6.3 % and 3.6 %, respectively, of the total accident rate on the inland waterways transport of the Russian Federation [1]. According to statistics overview annually is-sued by European Maritime Safety Agency (EMSA), over the 2014-2020 period, ship collisions represented 21.7 % of all casualty events [2]. When analyzing underlying factors leading to mari-time accidents, safety investigations determined that 70% of safety investigations were related to “Human Factor” [3].Enhancement of navigational safety in the world merchant fleet is carried out through a number of technical, organizational, economic, environmental and legal standards aimed at pre-venting the occurrence of casualty events, saving human life at sea and reducing environmental risks. In order to reduce or eliminate the need for human involvement in ship control systems and promote navigation safety in the maritime transport sector, Artificial Intelligence (AI) technology based on Machine Learning (ML) methods is being increasingly integrated into maritime transpor-tation industry to create and implement the technology of Unmanned Navigation (UN), providing a reduction of the ship's operational costs, reduction in the ship crew size or complete elimination of the ship crew and a decrease in environmental impacts from direct emissions.Recently, the technology of UN has become a widely discussed topic and one of the fastest growing field of maritime shipbuilding and marine robotics. Furthermore, techniques and meth-ods of ML effectively addressed challenges of ensuring safe remote control of transport vehicles and intellectualization of complex systems. Taking this background into consideration, the devel-opment of an algorithm for the autonomous vessel collision avoidance system (ACAS) in UN is a new trend in the maritime sector and coming out of the growing field of marine robotics.

About the authors

L. A. Barakat

Astrakhan State Technical University

I. Y. Kvyatkovskaya

Astrakhan State Technical University

References

  1. Анализ и состояние аварийности. [Электронный ресурс] / Министерство транспорта Российской Федерации. – URL: https://rostransnadzor.gov.ru/rostransnadzor/podrazdeleniya/sea/deyatelnost-podrazdeleniya/81 (дата обращения: 23.06.2023).
  2. Annual overview of marine casualties and incidents 2021. [Электронный ресурс] / European Maritime Safety Agency. – URL: https://www.emsa.europa.eu/publications/download/6955/4266/23.html (дата обращения: 23.06.2023).
  3. Galieriková A. The human factor and maritime safety/A. Galieriková// 2019 13th International Scientific Conference on Sustainable, Modern and Safe Transport (TRANSCOM 2019), High Tatras, Novy Smokovec – Grand Hotel Bellevue, Slovak Republic, May 29-31, 2019. – Transportation Research Procedia, 2019. – Vol. 40 – P. 1319–1326. doi: 10.1016/j.trpro.2019.07.183
  4. Developing a ship collision risk Index estimation model based on Dempster-Shafer theory/M. Abebe, Y. Noh, C. Seo, D. Kim, I. Lee // Applied Ocean Research. 2021. – Vol. 113. doi: 10.1016/j.apor.2021.102735
  5. Vessel Multi-Parametric Collision Avoidance Decision Model: Fuzzy Approach / T. Brcko, A. Androjna, J. Srše, R. Boć // Journal of Marine Science and Engineering. – 2021. – Vol. 9(1). – P. 49. doi: 10.3390/jmse9010049.
  6. Sun Z., Fan Y., Wang G. An Intelligent Algorithm for USVs Collision Avoidance Based on Deep Reinforcement Learning Approach with Navigation Characteristics // Journal of Marine Science and Engineering.2023. – Vol. 11(4). – P. 812. doi: 10.3390/jmse11040812
  7. A research on AIS-based embedded system for ship collision avoidance / D. Chen, C. Dai, X. Wan, J. Mou // 2015 The 3rd International Conference on Transportation Information and Safety (IC-TIS): proceedings, Wuhan, China, 25-28 June 2015. – IEEE, 2015. – P. 512–517. doi: 10.1109/ICTIS.2015.7232141
  8. Estimation of vessel collision risk index based on support vector machine / L. Gang, Y. Wang, Y. Sun, L. Zhou, M. Zhang // Advances in Mechanical Engineering. – 2016. – Vol. 8(11). doi: 10.1177/1687814016671250
  9. Park J., Jeong J.S. An estimation of ship collision risk based on relevance vector machine // Journal of Marine Science and Engineering. – 2021. – Vol. 9(5). – P. 538. DOI: 10.3390/ jmse9050538.
  10. О проведении эксперимента по опытной эксплуатации автономных судов под Государственным флагом Российской Федерации: Постановление Правительства Российской Федера-ции от 05.12.2020 № 2031 [Электронный ресурс] // Официальное опубликование правовых актов. – URL: http://publication.pravo.gov.ru/document/0001202012080021(дата обращения: 24.06.2023).
  11. Оценка рисков эксплуатации безэкипажных судов / А.В. Титов, Л. Баракат, О.Ю. Ла¬зовская, Г.А. Тактаров, О.П. Ковалев // Морские интеллектуальные технологии. – 2019. – Vol. 4, № 1(43). – C. 11–23.
  12. Баракат Л. Предотвращение столкновений безэкипажных судов с использованием глубокого обучения с подкреплением // Пятая международная научно-практическая конференция. «Имитационное и комплексное моделирование морской техники и морских транспортных систем» (ИКМ МТМТС-2019), Санкт-Петербург (Россия), 10 июля 2019 года. – М.: Перо, 2019. – С. 102–105.
  13. Баракат Л.А., Квятковская И.Ю. Интеллектуальное принятие решений по автономному предотвращению столкновений безэкипажных судов на основе алгоритма глубокой Q-сети // 64-я Международная научная конференция Астраханского государственного технического университета. Астрахань, 20–25 апреля 2020 года. – Астрахань: Изд-во АГТУ, 2020.
  14. Collision Avoidance Method for Autonomous Ships Based on Modified Velocity Obstacle and Collision Risk Index / K. Zhang, L. Huang, Y. He, L. Zhang, W. Huang, C. Xie, G. Hao // Journal of Advanced Transportation. 2022. – Vol. 2022. – 25 p. doi: 10.1155/2022/1534815
  15. Collision avoidance algorithm based on COLREGs for unmanned surface vehicle / H.G. Kim, S.J. Yun, Y.H. Choi, J.K. Ryu, J.H. Suh // Journal of Marine Science and Engineering. – 2021. – Vol. 9(8). – P. 863. doi: 10.3390/jmse9080863
  16. Simulation on local obstacle avoidance algorithm for unmanned surface vehicle / C. Wang, Y.S. Mao, K.J. Du, B.Q. Hu, L.F. Song // International Journal of Simulation Modelling. – 2016. – Vol. 15(3). – P. 460–472. doi: 10.2507/IJSIMM15(3)6.347
  17. McMillan M. Data structures and algorithms using Visual Basic. NET. – New York: Cambridge University Press, 2005. – 401 p.

Statistics

Views

Abstract - 34

PDF (English) - 26

Refbacks

  • There are currently no refbacks.

This website uses cookies

You consent to our cookies if you continue to use our website.

About Cookies