Traffic in megacities is an interesting problem. On paper its straightforward, yet in practice understanding how to effectively manage congestion is hard. Foreseeable solutions include slot-based traffic systems or, more innovatively, adaptive machine learning based systems in which the cities adapt to patterns instead of the reverse. Today, a majority of cities still run using an outdated variant of fixed-clock systems in which traffic lights switch with a predefined time scale based on timers set years ago, denying the system the ability to adapt to real-time events. We have a system in which congestion occurs, ambulances are forced to wait behind the congested traffic, and significant amounts of fuel are wasted idling, all while the lights are switching according to a set timer. This paper develops TrafficAI, a system in which cameras feed real time video analysis, vehicle detection, and counting to inform the timers of live, dynamic data in an effort to seek congestion reduction. To avoid overreacting, we parameterized the inputs using an Exponential Moving Average (EMA) approach to smooth out the timeseries. To mitigate robotic flickering, and using an hysteresis approach, we used a decision engine that deemed when to switch at an appropriate rate. Our system also implemented a Finite State Machine (FSM) to guarantee that state transitions occurred safely (transition events came in order with the colors). Finally, our system detected the presence of an emergency vehicle and activated a signal override to ensure the firetruck and ambulance can pass unencumbered. Our experiments demonstrated that we could achieve measurable improvements in traffic flow.
[1]Zhang et al., “Analysis of Vehicle and Pedestrian Detection Effects of Improved YOLOv8 Model in Drone-Assisted Urban Traffic Monitoring System,” PLOS ONE, vol. 20, no. 3, p. e0314817, Mar. 2025.
[2] Li et al., “Enhanced Multi-Target Detection in Complex Traffic Using an Improved YOLOv8 with SE Attention, DCN_C2f, and SIoU,” World Electric Vehicle Journal, vol. 15, no. 12, p. 586, Dec. 2024.
[3]M. Carrasco et al., “Traffic Sign Detection and Recognition Using YOLO Object Detection Algorithm: A Systematic Review,” Mathematics, vol. 12, no. 2, p. 297, Jan. 2024.
[4]R. Shah et al., “Machine Learning Solutions for Adaptive Traffic Signal Control: A Review of Image-Based Approaches,” World Journal of Advanced Engineering Technology and Sciences, vol. 13, no. 1, pp. 476–481, Sep. 2024.
[5]A. Polydoros et al., “Traffic Signal Control via Reinforcement Learning: A Review on Applications and Innovations,” Infrastructures, vol. 10, no. 5, p. 114, May 2025.
[6]Y. Li et al., “Incorporating Vision-Based Artificial Intelligence and Large Language Model for Smart Traffic Light Control,” Engineering Applications of Artificial Intelligence, Elsevier, 2025.
[7]A. Omari Alaoui et al., “Real-Time Traffic Signal Adjustment Using YOLOv8 for Improved Integration of Emergency Vehicles in Smart Traffic Systems,” Signal, Image and Video Processing, Springer Nature, 2025.
[8]C. Johny and A. Sharma, “Emergency Vehicle Detection for Signal Optimization Using YOLOv8,” Journal of Electrical Systems, vol. 20, no. 3, pp. 6952–6960, 2024.
[9]R. Putra et al., “A Systematic Review on Vision-Based Traffic Density Estimation for Intelligent Transportation Systems,” IET Intelligent Transport Systems, Wiley, 2025.
[10]M. Sankaranarayanan, C. Mala, and S. Jain, “Traffic Density Estimation for Traffic Management Applications Using Neural Networks,” International Journal of Intelligent Information Technologies, vol. 20, no. 1, pp. 1–19, 2024.
[11]V. Kumar et al., “Machine Learning Based Adaptive Traffic Prediction and Control Using Edge Impulse Platform,” Scientific Reports, Nature Portfolio, May 2025.
S. Abbas et al., “Vision Based Intelligent Traffic Light Management System Using Faster R-CNN,” CAAI Transactions on Intelligence Technology, IET/Wiley, Apr. 2024.
How to Cite This Paper
Dr. Aravindhan k, Y.Someswar Naidu, D.Abinav Datha, G.Chakravardhan Reddy, Y.M.Hanish Kumar Reddy (2026). Creating Intelligent Devices To Improve The Commutation Sector. International Journal of Computer Techniques, 13(2). ISSN: 2394-2231.
