Traffic Sign Recognition for Automated Speed Control Using Deep Learning*

Abstract

Today's newest and most popular topic is real-time driverless cars and their applications. To meet the requirements of these systems, complex algorithms must efficiently perform essential tasks. Autonomous driving technology presents one of the biggest challenges in speed control. We present a new method for speed instruction-based motor control that uses Convolutional Neural Networks (CNNs) to predict speed limits. CNNs extract crucial information from traffic sign signatures, and the performance of our system relies on its ability to accurately identify and classify objects in images. The Arduino microcontroller uses the output of the CNN to control the speed of the robot chassis. Experimental results, both in simulations and real-world tests, demonstrate the effectiveness of our approach. The proposed model achieves a high accuracy of 96% to 97% and outperforms existing models with a faster recognition time, maintaining the required speed with minimal error. These findings highlight the potential of CNN-based systems in enhancing autonomous driving technology.