Integrated Traffic Management System: A review
Year: 2024
Type: conference-paper
Abstract: Traffic congestion is a significant global issue, particularly in India's rapidly urbanizing cities. Traditional traffic management methods, such as static signal timings and manual enforcement, have proven insufficient in addressing the growing challenges of urban mobility. This paper presents a comprehensive review of research efforts aimed at improving traffic management through vehicle detection, violation detection, and dynamic signal control. Object detection technologies, particularly YOLO-based approaches, offer promising solutions for real-time traffic monitoring and violation detection. By deploying cameras equipped with these algorithms at critical intersections, authorities can monitor traffic flow, detect violations, and dynamically adjust signal timings. Several studies have investigated the integration of deep learning models such as YOLOv3 and YOLOv5 for precise vehicle detection and violation identification, showcasing high accuracy in vehicle counting, speed violation detection, and license plate recognition. Furthermore, the development of Autonomous Road Surveillance Systems (ARSS) and the integration of IoT-machine learning algorithms enable adaptive traffic management, resulting in reduced congestion and improved efficiency. This paper reviews advancements in integrated traffic management systems, focusing on dynamic traffic light control and adaptive signal control methods that utilize connected vehicle data to dynamically optimize traffic signals, reducing delays and queue lengths. Additionally, the study explores dynamic traffic management approaches, emphasizing the need for real-time adjustments based on traffic conditions.
Authors: Akshata Shanmugam, Celestin Fernandes, Dikshant Sharma, Vikas Upadhyaya
Institution: NIIT University
Language: English
Subfield: Control and Systems Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4403854086