Vision-Based Cyber-Physical Traffic Signal Control Using YOLOV8 Multi-Parameter Congestion Scoring With Arduino UNO And ESP32–MQTT Prototypes

Vehicles at urban crossings frequently linger at a red light while the crossing road is empty because fixed-time signal systems are unable to adjust to the way traffic demand varies throughout the day. We introduce a camera-based Cyber-Physical System (CPS) in this study that completes the loop between the signal head and the road scene. Traffic footage within an operator-selected region of interest (ROI) is processed by YOLOv8 and OpenCV. A weighted congestion score is created by combining three indicators—vehicle count, occupancy ratio, and queue depth—that are obtained from the detections. A guarded rule assigns a HIGH or LOW state and provides a microcontroller with the corresponding phase plan when the score, count, and queue all hit their respective criteria. Two hardware variations are powered by the same software pipeline: an ESP32 that subscribes to commands via MQTT and an Arduino UNO that is provided via a USB serial link. Infrared car sensors are not used in either version. In bench tests, a run with 12 detected vehicles, an occupancy ratio of 0.51, and a queue depth of 168.6 pixels yielded a score of 0.46, was marked HIGH, and created a plan on the LED signal head that was 20 seconds green, 2 seconds amber, and 10 seconds red. The project is now at the prototype stage; further work will include field tests as well as quantitative accuracy and latency assessments.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-18
DOI
https://doi.org/10.5281/zenodo.24337831
Primary Topic
Traffic control and management
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Vision-Based Cyber-Physical Traffic Signal Control Using YOLOV8 Multi-Parameter Congestion Scoring With Arduino UNO And ESP32–MQTT Prototypes

Astik Sarkar
Zenodo (CERN European Organization for Nuclear Research)
Traffic control and management
article

Vision-Based Cyber-Physical Traffic Signal Control Using YOLOV8 Multi-Parameter Congestion Scoring With Arduino UNO And ESP32–MQTT Prototypes

Astik Sarkar
article en

Abstract

Vehicles at urban crossings frequently linger at a red light while the crossing road is empty because fixed-time signal systems are unable to adjust to the way traffic demand varies throughout the day. We introduce a camera-based Cyber-Physical System (CPS) in this study that completes the loop between the signal head and the road scene. Traffic footage within an operator-selected region of interest (ROI) is processed by YOLOv8 and OpenCV. A weighted congestion score is created by combining three indicators—vehicle count, occupancy ratio, and queue depth—that are obtained from the detections. A guarded rule assigns a HIGH or LOW state and provides a microcontroller with the corresponding phase plan when the score, count, and queue all hit their respective criteria. Two hardware variations are powered by the same software pipeline: an ESP32 that subscribes to commands via MQTT and an Arduino UNO that is provided via a USB serial link. Infrared car sensors are not used in either version. In bench tests, a run with 12 detected vehicles, an occupancy ratio of 0.51, and a queue depth of 168.6 pixels yielded a score of 0.46, was marked HIGH, and created a plan on the LED signal head that was 20 seconds green, 2 seconds amber, and 10 seconds red. The project is now at the prototype stage; further work will include field tests as well as quantitative accuracy and latency assessments.

Zenodo (CERN European Organization for Nuclear Research)
Industry, innovation and infrastructure, Sustainable cities and communities
Traffic control and management
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon