Exploring diverse prediction models in intelligent traffic control

Abstract

Traffic congestion is a major challenge that affects excellence of life for numerous people across world. The fast growth in many vehicles contributes to congestion during peak and non-peak hours. The vehicle traffic resulted in many issues like accidents and inefficiency in traffic flow. Many traffic light control systems operate on fixed time intervals leads to inefficiency. The fixed-time signals cause unnecessary delays on roads with minimum number of quantity vehicles. Intelligent transport systems (ITS) introduce new comprehensive framework that combine the advanced technologies to improve the transportation network efficiency and to optimize the traffic management. The high-traffic routes are forced to wait excessively. Machine learning (ML) methods have designed to examine the traffic control. However, the accurate detection and vehicle tracking are essential one for effective ITS. In order to mention these problems, ML and deep learning (DL) methods are introduced to improve prediction performance.