Automatic berthing systems: a review on artificial intelligence methods for marine ship

Abstract

One of the most challenging issues in the world of ship control is automatic ship berthing. In order for the berthing to be completed automatically and safely, the control systems must account for the ship’s dynamics at low speeds. Artificial neural networks (ANNs) are frequently used to address this need because of their capacity to mimic and carry out all the functions of the human brain throughout the ship berthing procedure. However, there are still certain drawbacks when employing this theory to create the automatic system for ship berthing, and this makes it more challenging to develop the control system for practical ship applications. This study reviews the pros and cons of employing ANNs along with some other deep learning algorithms in automatic ship berthing systems. Some directions for further research into automatic ship-berthing systems are also suggested.