Optimizing Driver Identification Using Adaboost and Vehicular Telemetric Data Analysis

In Intelligent Transportation Systems (ITS), Artificial Intelligence (AI) has transformed transportation network management. This paper explores driver identification using machine learning to enhance vehicle security and efficiency. AI applications in ITS enable real-time data processing, predictive maintenance, and better traffic management, promoting a safer and more efficient ecosystem. Driver identification, or finger-printing, utilizes machine learning to identify drivers based on their unique driving behaviors, helping prevent unauthorized vehicle use, reducing theft, ensuring regulatory compliance, and optimizing resource allocation. Our research employs the AdaBoost algorithm, leveraging OBD- II data such as vehicle speed, RPM, coolant temperature, torque, accelerometer, and GPS for accurate driver identification with an accuracy of 96%. This approach capitalizes on AdaBoost's real-time processing capability, enhancing prediction accuracy, reducing overfitting, and improving generalization to new data. The results highlight the significant potential of AI -driven methods to advance ITS capabilities.