Preventing Road Crashes with IoT Monitoring of Driver Fatigue in Real-Time

Kashyap, Abhijan S.; Chowdhury, Sumit Singha; Hegde, Ananya V. · 2025 · Crossref

DOI: 10.36548/jismac.2025.3.001

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the critical public safety issue of road traffic accidents caused by driver fatigue, a factor responsible for significant fatalities globally and in India. While existing drowsiness detection systems often rely on camera-based computer vision or wearable physiological sensors, these solutions suffer from high costs, environmental sensitivity (particularly lighting conditions), user discomfort, and privacy concerns. To overcome these limitations, the authors propose a technically novel, non-intrusive, and low-cost Internet of Things (IoT) system designed for real-time monitoring of driver fatigue. The primary objective is to provide a scalable, energy-efficient solution that functions effectively in low-light conditions without requiring physical contact with the driver. The system architecture centers on an ESP32 microcontroller, which serves as the edge computing unit. It integrates two primary sensors: an infrared (IR) eye-blink sensor to detect prolonged eye closure and an MPU6050 Inertial Measurement Unit (IMU) to monitor abnormal head movements, such as nodding or tilting. The detection logic employs rule-based algorithms rather than complex machine learning models to ensure low latency and computational efficiency on embedded hardware. Specifically, the system flags drowsiness if eyes remain closed for more than five seconds or if head pitch deviations exceed ±15 degrees rapidly, or if the head tilts more than 20 degrees for over three seconds. Upon detecting these indicators, the system triggers immediate local alerts via a buzzer and LED. Simultaneously, the ESP32 transmits status data, including driver ID, sensor readings, and timestamps, to a Firebase Real-time Database via Wi-Fi. This data is visualized on a custom web dashboard, enabling remote monitoring by supervisors or authorities. The system was evaluated in a simulated driving environment with human participants to assess detection accuracy and alert responsiveness. The results demonstrated a detection accuracy of 90%, with alerts issued in under one second. Anecdotal feedback from users indicated that the dual-alert mechanism (auditory and visual) was effective in regaining driver attention. The study highlights that the hybrid IR-IMU sensing approach, combined with cloud integration, offers a robust alternative to camera-based systems, particularly for night-time operation. The significance of this work lies in its provision of a low-cost, scalable, and non-intrusive framework for enhancing road safety. By leveraging IoT and edge computing, the system enables not only immediate driver intervention but also remote oversight, allowing for proactive safety management in fleet operations or personal transport. The authors conclude that this approach addresses key gaps in current literature regarding cost, environmental robustness, and real-time remote monitoring capabilities, offering a practical solution for reducing fatigue-related crashes.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-nvidia on 2026-08-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• drowsiness detection algorithms
• dms validation

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: physiological data
• Methodological Resource: tool software, validation psychometrics