INTELLIGENT TRAFFIC CONGESTION CONTROL SYSTEM USING A CONTROLLED MACHINE LEARNING ALGORITHM ON ADAPTIVE IOTN

Abstract

The phenomenon of congestion on the roads occurs when the demand rate on the road or on
a transport facility exceeds the available capacity, and there are two types: either routine, i.e.
occurs at certain times that are peak, for example, on the road, walking or returning from work or
educational institutions of people; or another type – sudden traffic jams that have appeared as a
result of a traffic accident, that is, in the event of an accident on the road, or due to other force
majeure reasons. In this regard, in order to reduce the increase in congestion in cities, it is possible
and necessary to use the concept of smart systems in modern conditions of life and technology
development. It is distinguished by a variety of algorithms used in the world of machine learning(ML) and the Internet of Things (IoT) to more accurately predict the flow of traffic in the short
term and identify opportunities to prevent congestion. In modern cities, many different sensors can
be used to collect information to predict short-term traffic in the city and accurately capture the
spatial and temporal evolution (change) of traffic flow. Algorithms embedded in machine learning
improve the capabilities of the system being developed. The quality of the decisions made by the
developed artificial intelligence increases with a simultaneous increase in the volume of data collected.
This article proposes a model of the TCC-SVM system for analyzing traffic jams in a smart
city environment. The proposed model includes an Internet of Things (IoT) traffic management
system that reports congestion at a certain point. Existing traffic management systems are becoming
ineffective due to the increase in the number of vehicles on the roads. In urban areas, traffic
jams and accidents are a serious problem. An intelligent transport system is necessary to solve the
problems caused by congestion on the roads.