Machine Learning Method for Optimal Route Calculation

The industrialization of society and the transition to Industry 4.0 requires the use of advanced data processing and decision-making technologies. In connection with the utilization of perfect measuring systems and monitoring of a large number of parameters, the volume of statistical data increases when it is necessary to process them in real time. On the other hand, the computing possibilities of information systems is also increasing, which contributes to expanding the opportunity of using artificial intelligence technologies based on machine learning and deep learning methods. This paper is devoted to the development and analysis of a new machine learning method for determining the optimal route of an unmanned system or a guided vehicle. The initial information for building a route is a map with the designation of permitted and prohibited zones of movement from a given point to the final destination. The proposed method is an improvement of DBSCAN nonparametric clustering technologies and Dijkstra's algorithm.