An overview of application of artificial neural networks in modeling motions of autonomous vessel
EDN: UFKSYW
Abstract
The relevance of the chosen research topic is explained by the active development of autonomous navigation and the need to solve problems arising during the operation of marine autonomous surface ships.
This paper provides an overview of some models for predicting motions of a ship in irregular waves, built on the basis of artificial neural networks. The practical advantage of the considered approaches lies in the possibility of their application as part of relatively low-performance onboard computing systems of autonomous ships, where it is economically impractical or technically impossible to place high-performance computing clusters, and the calculation results should be obtained in real time. It is concluded that the calculation results obtained using the considered models have a significant error in predicting the ship's motions parameters near the extremes of the curve of the angles of rolling and pitching, which calls into question the safety of their application.
About the Authors
S. V. Shults
St. Petersburg State Marine Technical University
Russian Federation
Russian Federation
PhD student
190121; Lotsmanskaya ul., 3; St. Petersburg
М. А. Kuteinikov
FAI Russian Maritime Register of Shipping
Russian Federation
Russian Federation
DSc
191186; Millionnaya ul., 7A; St. Petersburg
V. Yu. Shults
St. Petersburg State Marine Technical University
Russian Federation
Russian Federation
PhD
190121; Lotsmanskaya ul., 3; St. Petersburg
References
1. Madusanka N.S., Fan Y., Yang S., Xiang X. Digital twin in the maritime domain : a review and emerging trends. Journal of Marine Science and Engineering. 2023. Vol. 11. Issue 5. P. 1021. DOI: 10.3390/jmse11051021.
2. Aizinov S.D., Butsanets A.A., Smolentsev S.V., Senchenko V.G., Lopatin M.S. Principles for assessing the functional properties of systems for autonomous shipping. Research Bulletin by Russian Maritime Register of Shipping. 2024. No. 74. P. 83 — 96. (In Russ.)
3. Kasyk L., Wolnowska A.E., Pleskacz K., Kapuściński T. The analysis of social and situational systems as components of human errors resulting in navigational accidents. Applied Sciences. 2023. Vol. 13. Issue. 11. P. 6780. DOI: 10.3390/app13116780.
4. Gomzyakov M.V. Overview of maritime transport casualties in the Far East. Research Bulletin by Russian Maritime Register of Shipping. 2020. No. 58/59. P. 4 — 10. (In Russ.)
5. Kazunin D.V., Efimov V.V. Development of draft RS requirements for autonomous vessels. Research Bulletin by Russian Maritime Register of Shipping. 2020. No. 60/61. P. 4 — 22. (In Russ.)
6. Kuteinikov M.A., Samoilov V.R. Criteria of risk of stability loss under dead ship condition. Research Bulletin by Russian Maritime Register of Shipping. 2019. No. 56/57. P. 40 — 50. (In Russ.)
7. Liu Y., Zheng Q., Duan W., Huang L. Improving deterministic pitch motions estimation using bivariate sequential wave input. IOP Conference Series: Materials Science and Engineering. 2019. Vol. 688. Issue 3. P. 033017. DOI: 10.1088/1757-899X/688/3/033017.
8. Suhermi N., Suhartono S., Prastyo D.D., Ali B. Roll motion prediction using a hybrid deep learning and ARIMA model. Procedia Computer Science. 2018. Vol. 144. P. 251 — 258. DOI: 10.1016/j.procs.2018.10.526.
9. Zhang W., Liu Zh. Real-time ship motion prediction based on time delay wavelet neural network. Journal of Applied Mathematics. 2014. Vol. 4. Article ID 176297. DOI: 10.1155/2014/176297.
10. Su Y., Lin J., Zhao D., Guo C. et al. Real-time prediction of large-scale ship model vertical acceleration based on recurrent neural network. Journal of Marine Science and Engineering. 2020. Vol. 8. Issue 10. P. 777. DOI: 10.3390/jmse8100777.
11. Tian X., Song Y. Machine learning for short-term prediction of ship motion combined with wave input. Applied Sciences. 2023. Vol. 13. Issue 9. P. 5298. DOI: 10.3390/app13095298.
12. Wang Y., Wang H., Zou D., Fu H. Ship roll prediction algorithm based on Bi-LSTM-TPA combined model. Journal of Marine Science and Engineering. 2021. Vol. 9. Issue 4. P. 387. DOI: 10.3390/jmse9040387.
Review
For citations:
Shults S.V., Kuteinikov М.А., Shults V.Yu. An overview of application of artificial neural networks in modeling motions of autonomous vessel. Research Bulletin by Russian Maritime Register of Shipping. 2025;(78):44-50. (In Russ.) EDN: UFKSYW
Views:
148
Email this article 