Smart traffic lights : an IoT-based adaptive traffic management system

Abstract

Traffic congestion has been a pressing issue in many developing countries, including Nigeria, due to outdated infrastructure and static traffic control systems. This thesis presents the design and implementation of an IoT based smart traffic light system aimed at improving real time traffic management through adaptive signal control. Using microcontrollers, sensors, and cloud computing, the system detected real time traffic density, pedestrian movement, and emergency vehicles to dynamically adjust traffic signals. The study integrated Passive Infrared (PIR) sensors, sound sensors, and ESP32 microcontrollers with the Blynk IoT platform for cloud based monitoring and control. The smart system was tested in a simulated environment, and its performance was benchmarked against conventional fixed timer traffic lights. Results from controlled simulations demonstrate that the smart system significantly outperforms conventional fixed timer systems. Specifically, the smart traffic light reduced vehicle wait times from an average of 18 seconds to 10 seconds, and decreased emergency vehicle signal response times from approximately 9 seconds to just 2 seconds. Sensor accuracy reached 96.4% for the PIR sensor and 94.7% for the sound sensor. Additionally, cloud based monitoring allowed for a dashboard update interval of 30 to 50 seconds, with data transmission delays under 5 seconds. Despite encountering challenges related to sensor calibration and network reliability, the study confirmed that IoT based traffic systems provided a more efficient, responsive, and scalable solution for urban traffic control. This thesis presents a practical model for future deployment in smart city infrastructures, supporting sustainable urban mobility through data driven, intelligent traffic management.