Possibilities of Using the Architecture of Neural Networks "Transformer" in Transport Engineering

The development of industry increasingly requires intelligent automation, while automation includes neural network technologies. This article discusses neural network technologies, the use of which can significantly improve the quality and automation. Today, many neural networks perform various highly specialized tasks, the greatest interest for the tasks of transport engineering may be the expansion of multimodality and versatility to solve various tasks. Computer vision is of the greatest interest, and NLP (Natural language processing) can also be included here. The scientific novelty of this work lies in the consideration of machine learning methods for solving many problems of transport engineering and the study of the possible future of these neural networks. The study examines modern architectures and their practical application in the field of transport engineering