Traffic Condition Classification Using IoT on Raden Inten II Road

Abstract

Unmonitored traffic conditions often hinder decision-making processes in traffic management, particularly on secondary roads. Jalan Raden Inten II in East Jakarta is one of the connecting routes with heavy traffic activity at certain times, yet no integrated data-based monitoring system is currently available. This study proposes an Internet of Things (IoT)-based traffic condition classification system to identify Clear, Normal, or Congested states based on vehicle counts and speed categorization. The system is designed using an ESP32 microcontroller, an HB100 sensor to detect vehicle speed, and two AJ-SR04M ultrasonic sensors to detect vehicle presence. Data on vehicle counts and the percentage of slow-moving vehicles are periodically transmitted to the ThingSpeak platform and processed using the Threshold-Based Classification method. The classification results are visualized on a dashboard-based website equipped with charts, traffic condition status, and notifications when consecutive congestion is detected. Testing was conducted using simulation data over a specific period. Qualitative validation was carried out by comparing the classification results with traffic indicators from Google Maps. The results show that the system can classify traffic conditions with a good degree of agreement with external references, although discrepancies occurred at certain times due to the limitations of simulated data. This research demonstrates that a simple IoT approach can provide an affordable and effective solution for monitoring and classifying traffic conditions, with potential for real-world implementation in future studies.