An Adaptive Traffic Light Control Scheme and its Implementationin WSN-Based ITS

Binbin Zhou, Jiannong Cao, Jingjing LiInternational Journal on Smart Sensing and Intelligent Systems 2013 · DOI

Abstract

We investigate the problem of adaptive control of traffic lights using real-time traffic information collected by a wireless sensor network (WSN). Previous studies mainly focused on optimizing the intervals of green lights in a fixed sequence of traffic lights, and ignored some traffic flow’s characteristics and special traffic circumstances. In this paper, an adaptive traffic light control scheme has been proposed, in which the sequence of traffic lights can be adjusted dynamically in accordance with the real time traffic detected, including traffic volume, waiting time and vehicle density. Subsequently, the optimal traffic light length can be determined according to the local traffic and predicted traffic data. Simulation results demonstrate that the proposed scheme can achieve much higher performance, in terms of throughput and average waiting time. We also implement the proposed scheme into our WSN-based ITS project, iSensNet, and the result shows that our scheme is effective and practical.

Introduction

Traffic congestion is a significant global issue, leading to wasted fuel, time, and degraded air quality. Traditional control systems, such as fixed-time or basic actuated methods, are often too rigid to handle the complexities of modern urban traffic. Fixed-time systems operate on pre-set intervals regardless of demand, while actuated systems change based on presence but typically follow a fixed sequence. This paper addresses these limitations by proposing an adaptive scheme that dynamically adjusts both the sequence and duration of traffic lights using real-time data from Wireless Sensor Networks (WSNs).

Methodology

The proposed system utilizes a WSN-based architecture to monitor and control traffic at intersections. The approach is divided into three primary phases:

  1. Real-time Traffic Detection: A network of 16 sensor nodes is deployed at each intersection (two per lane). Sensors labeled $Xy1$ are placed at the intersection, while $Xy2$ sensors are placed at a distance ($SensorDistance$) to predict vehicle arrivals.
  2. Green Light Sequence Determination: The system calculates a Green Light Demand (GLD) for 12 distinct traffic cases. The GLD is a weighted function of traffic volume, waiting time, "hunger level" (to ensure fairness), "blank" detection (identifying gaps in flow), and special circumstances like emergency vehicles.
  3. Light Length Determination: After selecting the optimal case, the system determines the green light duration ($G_{next}$) required to clear the detected queue.
Intersection model and the 12 traffic movement cases

The variables used in these calculations, such as total throughput (TP) and average waiting time (AV GWT), are defined in the system's notation table, which also specifies the sensor placement logic ($Xy1$ and $Xy2$).

Definitions of variables and parameters used in the control scheme

Results and Evaluation

Simulation results comparing the proposed scheme against Fixed-Time Control (FTC) and Actuated Traffic Control (ATC) show significant improvements. In high-congestion scenarios (volume-to-capacity ratios between 0.4 and 1.0), the adaptive scheme achieved a throughput approximately 1.6 times higher than traditional methods. Furthermore, average waiting times remained consistently low (under 10 seconds per vehicle) even as traffic volume increased, avoiding the exponential spikes seen in fixed-time systems. The scheme was also successfully implemented on the iSensNet testbed, demonstrating its practical feasibility in a laboratory environment using physical model vehicles.

Conclusion

This research presents a robust, adaptive traffic control system that leverages WSNs for dynamic sequencing and timing. By accounting for variables like vehicle "hunger" and traffic gaps ("blanks"), the system optimizes intersection efficiency beyond what is possible with rigid sequences. While the current model assumes constant vehicle speeds and does not yet account for pedestrians, it provides a strong foundation for more efficient urban traffic management. Future work will focus on incorporating dynamic speed detection and pedestrian crossing requirements.

Summary updated

Abstract

Abstract We investigate the problem of adaptive control of traffic lights using real-time traffic information collected by a wireless sensor network (WSN). Previous studies mainly focused on optimizing the intervals of green lights in a fixed sequence of traffic lights, and ignored some traffic flow’s characteristics and special traffic circumstances. In this paper, an adaptive traffic light control scheme has been proposed, in which the sequence of traffic lights can be adjusted dynamically in accordance with the real time traffic detected, including traffic volume, waiting time and vehicle density. Subsequently, the optimal traffic light length can be determined according to the local traffic and predicted traffic data. Simulation results demonstrate that the proposed scheme can achieve much higher performance, in terms of throughput and average waiting time. We also implement proposed scheme into our WSN-based ITS project, iSensNet, and the result shows that our scheme is effective and practical.
BibTeX
@article{zhou2013adaptive,
  author = {Binbin Zhou and Jiannong Cao and Jingjing Li},
  title = {{An Adaptive Traffic Light Control Scheme and its Implementationin WSN-Based ITS}},
  journal = {International Journal on Smart Sensing and Intelligent Systems},
  year = {2013},
  doi = {10.21307/ijssis-2017-604},
}
Figure