DEEP LEARNING IN THE CONTEXT OF ARTIFICIAL INTELLIGENCE IN MARINE NAVIGATION: DEVELOPMENT PROSPECTS

Authors

  • V.V. Golikov
  • K.O. Siniuta

Keywords:

deep learning, artificial intelligence, maritime navigation, automation, security

Abstract

The article provides an in-depth analysis of the prospects for applying deep learning technologies in the maritime navigation field. It is shown that deep learning enables the automation and significant improvement of processes such as: ship recognition and classification based on visual characteristics, detection of obstacles in the vessel's path, monitoring and analysis of data from onboard sensors, forecasting emergency situations, development of augmented reality systems to enhance navigation, creation of virtual simulators, accurate weather forecasting, automated control of vessel movement.

The literature review demonstrates that deep neural networks can significantly improve the quality of combining images from different sensors for ship detection, analyze large amounts of data from onboard sensors, develop effective augmented reality systems for navigation, create realistic virtual simulators, build accurate weather forecast models, implement automated vessel motion control based on deep learning.

The authors emphasize that further optimization of algorithms for specific tasks, integration of data from additional sensors, development of specialized deep neural network architectures adapted to the maritime environment specifics, as well as thorough testing of systems under real-life conditions, are crucial for successful implementation of deep learning technologies in maritime navigation.

Particular attention should be paid to cybersecurity issues and protection of control systems against hacking. It is also important to develop methods for effective model training on the limited data typical for maritime navigation.

In conclusion, integration of deep learning technologies opens up significant prospects for automation and improving efficiency and safety of maritime transport. Successful implementation requires joint efforts of specialists in artificial intelligence, data processing, marine engineering, and navigation.

 

Published

2024-03-18