Enhancing Heavy Vehicle Safety with Intelligent Driver Behaviour Monitoring via Multi-Sensor Fusion and Embedded Systems

This research presents the development of a sensor fusion-based driver assistance system (DAS) aimed at enhancing the safety of heavy vehicle operators. The system integrates an alcohol sensor, an ADXL 335 accelerometer sensor, and an eye blink detection sensor with an Arduino platform for real-time monitoring of driver behaviour. LabVIEW software was employed for data processing and visualization, providing a user-friendly interface to track driver alertness, speed, and alcohol consumption. A classification algorithm was implemented to categorize the driver's condition into various risk levels based on sensor inputs. Alerts were generated in real-time to notify the driver of potential hazards, such as drowsiness, intoxication, or speeding. Experimental trials demonstrated the system's effectiveness in accurately detecting dangerous driving behaviours and issuing timely warnings. The use of sensor fusion enhanced data reliability by cross-verifying multiple inputs, reducing false positives. The system proved to be robust and practical for implementation in heavy vehicle operations, contributing to road safety. Future enhancements may include additional physiological sensors, integration with vehicle control systems, and the practise of machine learning (ML) algorithms to improve classification accuracy. This research highlights the potential of sensor fusion technology in the development of advanced DAS, particularly in the transportation sector.