Smart Adaptive Traffic Signal with Emergency and Pedestrian

Traditional fixed-time traffic signals often fail to adapt to dynamic urban traffic, resulting in
congestion and delayed emergency response. This research proposes a Smart Adaptive Traffic
Signal System with emergency and pedestrian priority, utilizing a three-tier IoT-enabled
architecture. The Sensing Layer employs the YOLOv5 deep learning model on an edge
processing unit to detect real-time vehicle queues and identify priority events such as
emergency vehicles. This data is transmitted to the Control Layer, based on an ESP32
microcontroller, which executes Hierarchical Control Logic: Tier 1 (Preemption Logic)
immediately grants right-of-way to critical vehicles, while Tier 2 (Adaptive Scheduling)
employs a Priority Queue Scheduling Algorithm (PQSA) to select the busiest lane.
References:
The Dynamic Time Scaling Algorithm (DTSA) calculates green light duration proportionally
to the detected vehicle queue, maximizing traffic flow. Inter-lane IoT communication enables
network-wide coordination. Simulation results demonstrate a 55.4% reduction in Average
Vehicle Waiting Time (AVWT) and a 20% increase in total throughput compared to fixed-time
systems under high-demand conditions. This intelligent traffic management approach enhances
urban mobility, reduces delays, and improves emergency response efficiency.