AI-Driven Traffic Sign Recognition and Speed Control for Autonomous Vehicle

Udhayakumar M, Govindaraju P, Ravichandran R, Arun V, Dhachinamoorthi S, Kamlesh Kannan G

2025

Abstract

Aim: This work suggests an AI-based traffic sign detection and speed control system in autonomous vehicles using YOLOv5 and PID controllers. The YOLOv5 model, implemented on Python using PyTorch, was trained on a pre-processed dataset following contrast stretching, noise removal, and rotation for improved generalization. Materials and Methods: Resizing of images was done to 640×640 pixels, and real-time detection using a 1080p camera attached to a vehicle. Efficient processing was handled by the platform using an Intel Core i7 10th Gen processor paired with an NVIDIA Jepson Niño. Compared to the conventional multi-stage CNN-based models, YOLOv5 enabled real-time detection at an inference time of 24.6 ms per frame. Result: The PID controller ensured smooth speed transitions according to observed traffic signs. Experimental results confirmed that YOLOv5 achieved an accuracy of 96.4% compared to 92.1% for conventional methods, with a lower false positive rate of 1.8% compared to 3.5%. The speed control system also attained a response accuracy of 98.5%, thus ensuring precise speed regulation. Conclusion: The above outcomes guarantee that YOLOv5, combined with PID controllers, significantly improves traffic sign detection and speed regulation and thus forms a practical solution for real-time autonomous vehicle implementation.

Paper Citation

in Harvard Style

M U., P G., R R., V A., S D. and G K. (2025). AI-Driven Traffic Sign Recognition and Speed Control for Autonomous Vehicle. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 133-137. DOI: 10.5220/0013923900004919

in Bibtex Style

@conference{icrdicct`2525,
author={Udhayakumar M and Govindaraju P and Ravichandran R and Arun V and Dhachinamoorthi S and Kamlesh Kannan G},
title={AI-Driven Traffic Sign Recognition and Speed Control for Autonomous Vehicle},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25},
year={2025},
pages={133-137},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013923900004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25
TI - AI-Driven Traffic Sign Recognition and Speed Control for Autonomous Vehicle
SN - 978-989-758-777-1
AU - M U.
AU - P G.
AU - R R.
AU - V A.
AU - S D.
AU - G K.
PY - 2025
SP - 133
EP - 137
DO - 10.5220/0013923900004919
PB - SciTePress