Deep Learning-Based Speed Limit Sign Detection Using YOLOv11 Applied to Speed Regulation in Electric Vehicles for ADAS

Abstract

Road safety and energy efficiency remain critical challenges in modern Electric Vehicles (EVs), particularly when drivers fail to adhere to speed limits. This study presents an effective speed limit sign detection and automatic speed regulation system using YOLOv11 within an Advanced Driver Assistance System (ADAS) framework. By integrating rapid sign detection with vector control of Permanent Magnet Synchronous Motors (PMSM), the proposed system delivers real-time speed limit compliance and improved vehicle performance. The YOLOv11 model was trained on a dataset of 23,000 traffic sign images. Experimental results demonstrate high performance, with a mean Average Precision (mAP) of 99.6% (mAP@50) and 86.2% (mAP@50–95), alongside 99.2% precision and 98.5% recall, underscoring the model's effectiveness. This work concludes that combining deep learning–based traffic sign recognition with advanced motor control significantly enhances ADAS capabilities and paves the way for future research into integrated, high-accuracy solutions for sustainable transportation.