Development of the Optimal Traffic Light Control Programs Using Machine Learning Technologies

One of the key directions in the development of intelligent transport networks is the introduction of automated traffic management systems. In the context of these systems, special attention is paid to the effective management of traffic lights, which are an important element of automated traffic management systems. The development of an automated system aimed at compiling an optimal program of traffic light signals on a certain section of the road network is presented. The development of an application that includes modules for modeling traffic flows, machine learning, and optimization of traffic light parameters is described. The Simulation of Urban Mobility (SUMO) traffic simulation package was chosen as a modeling tool, the BFGS (Broyden – Fletcher – Goldfarb – Shanno) optimization algorithm was used, and gradient boosting was used as a machine learning method. The results of practical research show that the developed system is capable of quickly and effectively optimizing the parameters of phases and duration of traffic light cycles, which significantly improves traffic management on the corresponding section of the road network. The proposed approach allows solving the problems of traffic flow management in conditions of limited availability of data on the road situation, which makes it promising for practical application, since currently for adaptive traffic light control systems the presence of expensive sensors and equipment is required, which is often impossible in real urban conditions.