This study presents a self-driving car system model based on smart image processing and machine learning (ML) algorithms to improve the recognition and response to traffic signs in a real-time environment. A combination of Arduino Uno and Raspberry Pi microcontrollers provides a cost-effective and high-performance solution. The system model includes detection algorithms for road properties i.e., lanes, objects, stop signs, and U-turn signs to improve the overall self-driving car system performance. An experimental test is conducted to evaluate the algorithm performance under varying operating conditions and the results were compared with those obtained from simulation. The outcomes of the experiments indicate that the system model achieved high accuracy in recognizing traffic signs and making decisions in real time, thus ensuring the safety of passengers and other road users.
Machine Learning Based Real-Time Self-Driving Car for Sudanese Autonomous Vehicular Markets
2024-05-19
594965 byte
Conference paper
Electronic Resource
English
Deep Learning Based Autonomous Driving in Vehicular Networks
| Springer Verlag | 2020
Sense4FL: Vehicular Crowdsensing Enhanced Federated Learning for Autonomous Driving
| ArXiv | 2025
Vehicular Networks and Autonomous Driving Cars
| Springer Verlag | 2021