An Innovative Framework for Traffic Light Control Based on the Combined Use of Artificial Intelligence and Multi-criteria Decision Making

XVII International Conference on Systems, Automatic Control and Measurements, SAUM 2024 (pp. 88-91)

АУТОР(И) / AUTHOR(S): Bratislav Lukić, Goran Petrović

Download Full Pdf  

DOI:  10.46793/SAUM24.088P

САЖЕТАК / ABSTRACT:

The concept of continuous development of smart cities focuses on the principle of interconnection, digitization and automation. In this context, the ubiquitous availability of various data has enabled the application of artificial intelligence and multi-criteria decision-making in various fields. The proposed model presents the management of traffic flows at the level of individual isolated signalized intersections, where the influence of traffic and non-traffic criteria was considered. Solving the problem in question is realized by developing a model for determining the optimal value of the length of the cycle and the distribution of green time and consists of four phases. The first phase represents the prediction of traffic flow using machine learning (ML), which is performed on the basis of historical – statistical data and regularities that occur between traffic flow, time intervals, temperature and weather conditions, as well as establishing the correlation between the mentioned traffic flow parameters. In the second phase, after obtaining the traffic flow, the length of the cycle and the distribution of green time are determined using a metaheuristic optimization method – a genetic algorithm, based on the minimization of vehicle time losses. The third phase represents the collection of data on traffic in real time, especially the encounter of priority vehicles using computer vision and public transport vehicles using information and communication technologies (ICT). In the last phase, the selection of the optimal value of the length of the cycle and the distribution of green time is made based on multi-criteria decision-making (MCDM). After choosing the optimal length of the cycle and the distribution of green time, the controller is activated at the intersection and traffic management is performed at the individual intersection.

КЉУЧНЕ РЕЧИ / KEYWORDS:

machine learning, genetic algorithm, MCDM, traffic light control

ЛИТЕРАТУРА / REFERENCE

  1. He, F.F., Yan, X.D., Liu, Y., Ma, L. A., „Traffic Congestion Assessment Method for Urban Road Networks Based on Speed Performance Index.”, Elsevier, Procedia Engineering, 137, pp. 425-433, 2016
  2. Chen, D., Yan, X., Liu, X.; Wang, L., Li, F., Li, S., „Multi-Task Fusion Deep Learning Model for Short-Term Intersection Operation Performance Forecasting“, Remote Sens. 13, pp. 1919, 2021.
  3. Kang, W. Zhao, B. Qi, and S. Banerjee, „Augmenting Self-Driving with Remote Control: Challenges and Directions,” Proceedings of the 19th International Workshop on Mobile Computing Systems & Applications, Tempe Arizona USA: ACM, Feb. pp. 19–24, 2018.
  4. Zhiqiang and L. Jun, „A review of object detection based on convolutional neural network,” 36th Chinese Control Conference (CCC), Dalian, China: IEEE, pp. 11104–11109, Jul. 2017.
  5. Chen, Y. Qiao, and Y. Li, „Inception-SSD: An improved single shot detector for vehicle detection,” J. Ambient Intell. Humaniz. Comput., vol. 13, no. 11, pp. 5047–5053, Nov. 2022.
  6. Bhattacharya and M. R. Virkler, „Optimization for Pedestrian and Vehicular Delay in a Signal Network,” Transp. Res. Rec. J. Transp. Res. Board, vol. 1939, no. 1, pp. 115–122, Jan. 2005.
  7. M. Madrigal Arteaga, J. R. Pérez Cruz, A. Hurtado-Beltrán, and J. Trumpold, „Efficient Intersection Management Based on an Adaptive Fuzzy-Logic Traffic Signal,” Appl. Sci., vol. 12, no. 12, p. 6024, Jun. 2022.
  8. J. Calle-Laguna, J. Du, and H. A. Rakha, Computing optimum traffic signal cycle length considering vehicle delay and fuel consumption. Transportation Research Interdisciplinary Perspectives, 3, 100021, 2019.
  9. Steadman and B. Huntsman, Connected vehicle infrastructure: Deployment and funding overview, No. PRC, 17-77 F, 2018.
  10. T. Truong, G. Currie, M. Wallace, C. De Gruyter, and K. An, “Coordinated Transit Signal Priority Model Considering Stochastic Bus Arrival Time,” IEEE Trans. Intell. Transp. Syst., vol. 20, no. 4, pp. 1269–1277, Apr. 2019.
  11. -H. Lee and H.-C. Wang, “A Person-Based Adaptive Traffic Signal Control Method with Cooperative Transit Signal Priority,” J. Adv. Transp., vol. 2022, pp. 1–17, Mar. 2022.
  12. Int Panis, S. Broekx, and R. Liu, “Modelling instantaneous traffic emission and the influence of traffic speed limits,” Sci. Total Environ., vol. 371, no. 1, pp. 270–285, Dec. 2006.
  13. E. Ukpebor, E. W. Omagamre, A. Bamidele, C. A. Unuigbe, E. N. Dibie, and E. E. Ukpebor, “Impacts of improved traffic control measures on air quality and noise level in Benin City, Nigeria”, Malawi J. Sci. Technol., vol. 13, no. 2, Art. no. 2, Dec. 2021.
  14. StoIlova and T. StoIlov, “Traffic noise and traffic light control”, Transp. Res. Part Transp. Environ., vol. 3, no. 6, pp. 399–417, Nov. 1998.
  15. Bendtsen and H. J. E. Larsen, Traffic management and noise. Road Directorate, Danish Road Institute, 2007.