Adaptive Traffic Signal Control Based on Forward-Searching Dynamic Programming Using Stop-Line Data

Abstract: Traffic signal control is a critical component in managing urban congestion and enhancing safety at intersections. Adaptive traffic signal control systems can adjust to the dynamic nature of traffic flow, thereby reducing delays and addressing inefficiencies associated with fixed-time signal control systems. In this context, various studies have explored the concept of dynamic programming as a key approach to enhancing the performance of adaptive traffic signal control systems. The main challenge of such systems is the accurate prediction of demand in the context of varying and complex traffic conditions, particularly when sensors placed upstream of the intersections fail to predict turning movements accurately. This paper presents a generalized adaptive traffic signal control (ASC) system that utilizes a forward-searching dynamic programming (FSDP) algorithm enhanced by stop-line detector data. The study demonstrates the effectiveness of this approach in improving traffic signal performance by leveraging stop-line detector data to achieve more accurate predictions of turning movements and overall traffic demand.
The proposed ASC framework employs real-time stop-line detection to gather data on vehicle arrivals and turning movements, allowing for a more precise assessment of traffic conditions. By analyzing this data, the system can dynamically adjust signal timings, thereby minimizing average delays per vehicle and improving overall intersection efficiency. A key focus of the study is the optimization of the planning period, a critical factor influencing the effectiveness of adaptive control strategies. Various planning periods were examined to determine the optimal duration that maximizes traffic flow and minimizes delays.
To evaluate the effectiveness of the proposed approach, simulation studies were conducted at isolated four-legged intersections using VISSIM, a traffic simulation software known for its ability to model complex traffic scenarios. This allowed for a detailed comparison of the FSDP-based ASC system against traditional fixed-time controls and previous dynamic programming methods. The results indicate that the implementation of stop-line data significantly enhances the accuracy of traffic signal planning, particularly in complex traffic scenarios characterized by asymmetric vehicle movements. Turning movement.
Simulation findings reveal substantial improvements in performance metrics, including reductions in average vehicle delays and queue lengths, particularly under asymmetric traffic conditions. This indicates that the integration of stop-line data into adaptive signal control systems yields better insights into real-time traffic dynamics, allowing for more responsive and efficient signal operations. Moreover, the study highlights the importance of carefully calibrating the planning period to ensure optimal system performance.
In conclusion, this research advances the field of traffic signal control by providing a robust framework for developing adaptive systems that can respond to real-time traffic conditions. This study can be extended to include the development of algorithms capable of dynamically adjusting the planning period in response to changing traffic patterns and extending this adaptive approach to coordinated traffic signal systems. By enhancing the adaptability of traffic control strategies, the findings of this study contribute to the ongoing efforts to improve urban traffic management and foster more sustainable transportation networks.