The use of artificial intelligence of unmanned vessels to determine obstacles when swimming

Abstract

Today, human activity increasingly depends on the ability to effectively use information. A modern specialist of any profile must be able to receive, process and use information using computers and other technical and mobile means and devices. Currently, a kind of technical revolution is taking place, related to the penetration of advanced information technologies such as Big Data, Internet of Things and Blockchain into various areas of our lives, and even into the maritime industry, which has traditionally lagged behind other industries in the application of the latest IT developments . Already today, the implementation of Artificial Intelligence (Artificial Intelligence) in navigation and control of the movement of the vessel is taking place. One of the most promising and possible for use on unmanned vessels is the method of deep learning of neural networks, which uses the "end-to-end learning" algorithm, capable of obtaining knowledge obtained through experience and the use of controllers to minimize the error when modeling the navigation map of the movement of an unmanned vessel. The authors of this work considered the methods of data recognition in shipping to build the trajectory of an unmanned vessel with artificial intelligence. In order to prevent collisions and build an accurate movement trajectory without additional fluctuations, the authors proposed the use of artificial intelligence with motion controllers with deep neural networks to identify various vessels using pattern recognition. It is shown that machine learning algorithms are capable of making intelligent decisions, but they can be complicated for unstructured data. These problems can be solved with the help of deep learning networks, in which a complex situation is solved using a multi-level hierarchical approach.