Raspberry Pi–based cyber-physical system for driver monitoring

Chee, Hao Jie (2024) Raspberry Pi–based cyber-physical system for driver monitoring. Final Year Project, UTAR.

PDF Download (4Mb)

Abstract

The revolution of Internet of Things (IoT) has transformed the way humans interacts with technology. This concept involves connecting devices via the internet, which allow for more efficient real-time communication and interaction regardless of distance. In the automotive field, the integration of IoT devices plays an important role in improving driver experience and safety. Nowadays, driver drowsiness and drunkenness are the two main causes of accidents worldwide. However, many existing solutions address these problems separately and leave gaps in protection. In address of this problem, a more comprehensive IoT-based driver monitoring system by using Raspberry Pi will be presented through this report. The proposed system combines several safety features into a single platform. For driver authentication, face embedding techniques is applied to verify the vehicle owner. Drowsiness detection is analysed through four criteria including eye landmarks, mouth aspect ratio, nose length ratio, and supported by a pretrained model to improve accuracy. Lastly, alcohol impairment is monitored through a gas sensor that measures the alcohol concentration in the air. The proposed system is also integrated with a buzzer, where it is activated upon the detection of any signs of fatigue or drunkenness to alert the driver immediately. To make the system more practical, a mobile application is developed to provide a simple and user-friendly interface. This application serve as a medium to display the output of the system where users are able to monitor the driver status, authentication results, and alerts in real time. This system aims to showcase how IoT hardware and mobile technology can be integrated to address multiple road safety issues at once and enhance driver safety.

Actions (login required)