Intelligent Traffic Congestion Classification System using Artificial Neural Network

Managing the ever increasing road traffic congestion due to enormous vehicular growth is a big concern all over the world. Tremendous air pollution, loss of valuable time and money are the common consequences of traffic congestion in urban areas. IoT based Intelligent Transportation System (ITS) can help in managing the road traffic congestion in an efficient way. Estimation and classification of the traffic congestion state of different road segments is one of the important aspects of intelligent traffic management. Traffic congestion state recognition of different road segments helps the traffic management authority to optimize the traffic regulation of a transportation system. The commuters can also decide their best possible route to the destination based on traffic congestion state of different road segments. This paper aims to estimate and classify the traffic congestion state of different road segments within a city by analyzing the road traffic data captured by in-road stationary sensors. The Artificial Neural Network (ANN) based system is used to classify traffic congestion states. Based on traffic congestion status, ITS will automatically update the traffic regulations like, changing the queue length in traffic signal, suggesting alternate routes. It also helps the government to device policies regarding construction of flyover/alternate route for better traffic management.