Study on simulation testbed development for traffic signal control application

• Main Author: Lin, Yulong
• Other Authors: Su Rong
• Format: Thesis-Master by Coursework
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Engineering
• Online Access: https://hdl.handle.net/10356/181201
• Institution: Nanyang Technological University

In urban environments, managing traffic efficiently remains a paramount challenge due to increasing vehicle numbers and limited infrastructure expansion possibilities. This dissertation develops a traffic signal control method, addressing the critical need for enhanced traffic management strategies. The aim of this study is to, first, conduct a horizontal comparative study of various control systems, evaluate the simulation results, and develop a more effective traffic signal control method that can adapt dynamically to changing traffic conditions and improve overall traffic flow efficiency in urban networks. The proposed method combines the benefits of Sydney Coordinated Adaptive Traffic System (SCATS) and Self-Organizing Traffic Lights (SOTL). Extensive simulation tests were conducted using the SUMO traffic simulation tool to evaluate the method under a peak hour traffic scenario, with a horizontal comparison by selecting various existing control methods, to demonstrate the advancements of the proposed method. The proposed method showed a considerable reduction in average vehicle delay and improved traffic throughput compared to traditional control methods. During peak hour traffic conditions, the hybrid method reduced the average delay by up to 7.5% compared to a conventional adaptive signal control scheme. These results were substantiated through rigorous sensitivity analyses, confirming the robustness and adaptability of the proposed approach. The comparative evaluation further highlighted the superior performance of the hybrid method over existing traffic control strategies. The research confirms that integrating SCATS with SOTL-like adaptations provides a more flexible and efficient traffic management solution. This hybrid method supports quicker response times to traffic variations and enhances the capability of traffic systems to manage congestion effectively, promising significant implications for urban traffic control strategies.