Machine Learning Prediction of Blood Alcohol Concentration: Drink and Drive Detection

The increasing in the number of road accidents produced by drunk and driving has led to substantial loss of life, emphasizing the vital requirement for an effective prevention system. This research presents the design and development of an intelligent alcohol detection and vehicle control system intended at mitigating such dangers and enhancing road safety. The system uses an MQ-3 alcohol sensor to detect the concentration of alcohol in a driver’s breath. As the concertation of alcohol, in the breath exceeds a predefined level, the system autonomously halts the vehicle and dispatches an alert containing the vehicle’s GPS coordinates via SMS to preconfigured emergency contacts. The system architecture is built around an Arduino Uno microcontroller, which interfaces seamlessly with peripheral components, including GSM, GPS modules, and motor drivers for vehicle control. A performance evaluation process demonstrates high detection accuracy using random forest machine learning algorithm, rapid response times, and efficient power consumption. The results validate the system’s efficacy in effectively stopping vehicle operation under the influence of alcohol, offering a viable and practical solution for enhancing road safety.