Evaluating the Performance of Intelligent Traffic Signals Based on Firefly Algorithm Applied in an Adaptive Control System

Subject Areas : Multimedia Processing, Communications Systems, Intelligent Systems

fariba jabbari ¹ ( MSc. Student, Civil Engineering (R&T), Yazd University, Yazd, Iran )
Mehdi Fallah Tafti ² ( Associate Professor, Department of Civil Engineering, Faculty of Engineering Yazd University, Yazd, Iran )

Received: 2022-08-02 Accepted : 2022-11-17 Published : 2022-09-23

Keywords: Adaptive Traffic Signals, Pre-time Traffic Signals, Firefly Algorithm, Traffic Signals Optimization,

Abstract :

IntroductionIn this article, the performance of adaptive traffic signal control systems based on an artificial intelligence technique, namely Firefly Algorithm, for traffic control at urban intersections has been investigated. In order to check the proposed algorithm, the required data were first collected from two intersections in Yazd city. These intersections were then simulated using AIMSUN traffic simulator software and calibrated and validated under existing conditions. In the next step, the adaptive control system based on the proposed Firefly algorithm was developed and then used in simulated intersections and its performance was compared with pre-time control system in terms of intersection traffic capacity and vehicle queue length at the entry approaches. The t-test was used for a more scientific investigation. For this reason SPSS software was used as one of the most widely used statistical software to perform this test.MethodIn order to concurrently optimize vehicle departure volume or throughputat and queue length at the entry approaches of each intersection, the Firefly algorithm was used to to develop a multi-objective adaptive traffic signal control logic and appropriately distribute the effective green time in each cycle between the entry approaches. ResultsThe simulation results indicated a lower average queue length and higher throughput when the proposed adaptive model was compared with pre-time model at both intersections. The t-test results showed that the adaptive traffic signal control method has resulted in a significant lower average queue length than the pre-time time control at one of the intersections with p-value equal to 0.002. However, this improvement was not statistically significant for the other intersection. Moreover, the t-test results on the average flow departure volume measure for both intersections indicated a significant improvement with p-value equal to 0.000. when the proposed adaptive method was compared to the pre- time method.DiscussionAccording to the results, the proposed adaptive model showed better overall performance in the scope of this research than the pre-time control method. The results indicate that the performance of adpative signal controls could be enhanced when artificial intelligence techniques such as Firefly algorithm are used in the control logic and a multi-objective optimization approach is used.

References:

 

• [1] Ebadi Shieuyari, Arman, Yazdanpanah, Shariat Mohaymeni, Abedini, Khashaeipur and Kalantari, " Calibrating traffic software based on the traffic conditions of Tehran city", Avaye Fahim, Third Volume: Providing instructions on how to simulate, calibrate and validate Imsan software, 1392.


• [2] Alvaro J. Calle-Laguna, Jianhe Du, Hesham A. Rakha, "Computing optimum traffic signal cycle length considering vehicle delay and fuel consumption", pp. 1-9, 2019.


• [3] Yan Li, Lijie Yu, Siran Tao, Kuanmin Chen, "Multi-Objective Optimization of Traffic Signal Timing for Oversaturated Intersection", 1-9, 2013.


• [4] Aleksandar Stevanovic, Jelka Stevanovic, Jaehyun So, Marija Ostojic, "Multi-criteria optimization of traffic signals: Mobility, safety, and environment", 1-23, 2015.


• [5] Wei Hao, Changxi Ma, Bahman Moghimi, Yuanyuan Fan, Zhibo Gao, "Robust Optimization of Signal Control Parameters for Unsaturated Intersection Based on Tabu Search-Artificial Bee Colony Algorithm", IEEE Advancing Technology for Humanity, pp. 32015-32022, 2018.


• [6] Hongluan Zhao, Guoyong Han, Xiaofei Niu, "The Signal Control Optimization of Road Intersections with Slow Traffic Based on Improved PSO", Mobile Networks and Applications, vol. 9, 2019.


• [7] Dex R.Aleko, Soufiene Djahel, "An Efficient Adaptive Traffic Light Control System for Urban Road Traffic Congestion Reduction in Smart Cities", pp. 1-20, 2020.


• [8] Nouha Rida, Mohammed Ouadoud, Abderrahim Hasbi. "Ant Colony Optimization for Real Time Traffic Lights Control on a Single Intersection", International Journal of Interactive Mobile Technologies (iJIM). pp. 196-214,


• [9] Bernardo Hubermen, Baldo Faieta, "Firefly: A Synchronization Strategy for Urban Traffic Control", 1-21, 2015.


• [10] Roberta Di Blasi, Arterial Signal Optimization through Traffic Microscopic Simulation, Sapienza University, Roma La Sapienza, Master degree in Transport systems engineering. pp. 1-90, 2016-2017.


• [11] Sahar Araghi, Abbas Khosravi, Douglas Creighton, "A Review on Computational Intelligence Methods for Controlling Traffic Signal Timing", Expert Systems with Applications, pp. 1538–1550, 2015.