Driver Drowsiness Monitoring System (DMS)

Overview

The Driver Drowsiness Monitoring System (DMS) is an IoT-based solution designed to enhance road safety by detecting driver drowsiness in real-time. The system uses hardware components such as the ESP32 microcontroller, ESP32-CAM module, and GPS-NEO-6M module, alongside cloud services like Firebase and ThingSpeak for data storage and visualization. The system captures driver face images and vehicle location to detect drowsiness, and it provides real-time monitoring via a web dashboard.

This project was completed as part of the PUSL2022 - Introduction to IoT module at the University of Plymouth.

Key Components

• ESP32 Microcontroller: Central control unit for coordinating modules.
• ESP32 CAM Module: Captures driver face images for drowsiness detection.
• GPS-NEO-6M Module: Tracks vehicle location and speed.
• Firebase Realtime Database: Logs vehicle data and drowsiness events.
• ThingSpeak: Cloud platform for real-time data visualization.
• Web Interface: Dashboard for visualizing real-time data.

System Architecture

1. Hardware:

   • ESP32 Microcontroller
   • ESP32 CAM Module
   • GPS-NEO-6M Module

2. Software:

   • Firebase for backend management
   • ThingSpeak for real-time monitoring
   • Web-based dashboard with HTML, CSS, and JavaScript

3. Data Flow:

   • Data is captured from sensors (GPS, camera) by the ESP32.
   • Drowsiness detection is performed locally on the ESP32-CAM images.
   • Data is logged to Firebase and visualized in ThingSpeak.

Database Design

• Schema:
  • drowsiness_events: Stores event details.
  • Child Nodes: Each event contains:
    • latitude: Vehicle’s latitude.
    • longitude: Vehicle’s longitude.
    • speed: Vehicle’s speed.
    • timestamp: Event time.

Deployment

• Hybrid Hosting: Combines on-premises servers and cloud services (Firebase, ThingSpeak) for real-time data storage and visualization.
• Infrastructure: Includes servers, virtualization for resource optimization, and seamless networking.

Testing

• Unit Testing: Validates individual components (ESP32, sensors, backend).
• Integration Testing: Verifies interactions between hardware, cloud services, and dashboard.
• System Testing: Ensures end-to-end functionality.

Dependencies

• Libraries:
  • OpenCV, dlib for facial detection and drowsiness monitoring.
  • Firebase Admin SDK, ThingSpeak API for cloud interaction.
  • Arduino libraries for Wi-Fi, HTTPClient, and sensor control.
• External Files:
  • Model files, alarm sounds, HTML for dashboard, Firebase credentials.

Communication Protocols

• Wi-Fi & HTTP: Enables communication between hardware and the system.
• Real-time Data Streaming: Streams sensor data for immediate action.

Contributors

• Sinel Nemsara
• Thejan Rajapaksha
• Yohan Nanayakkara
• Sachitha Eshan
• Charith Bandara
• Devin Fernando

Conclusion

The DMS effectively integrates hardware and cloud technologies to monitor driver behavior, detect drowsiness, and improve road safety through real-time monitoring and data visualization.