Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{194207,
        author = {MIDDE NANDINI and MOHAMMED SHOEB and MARADANI PRIYANKA and KUKKALA ANUDEEP and Ms B.REKHA},
        title = {AI-Powered Real-Time Drowsiness Detection and Auto Braking},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {10},
        pages = {2741-2745},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=194207},
        abstract = {Driver fatigue and drowsiness remain leading contributors to global road accidents, necessitating the development of intelligent in-vehicle safety systems. This research presents the design and implementation of an AI-powered real-time drowsiness detection and autonomous braking system designed to mitigate collision risks. The proposed system utilizes a hybrid approach, integrating computer vision algorithms with a hardware-based intervention mechanism. The software layer, developed in Python, leverages the OpenCV library for real-time video processing, while NumPy and SciPy are employed for mathematical modeling and calculating the Eye Aspect Ratio (EAR) and Mouth Aspect Ratio (MAR). By utilizing a 68-point facial landmark detector, the system identifies micro-sleep patterns and frequent yawning with high precision.
Upon detecting a drowsiness threshold, the software communicates via serial interface with an Arduino Nano microcontroller to initiate a multi-staged alert and safety protocol. The hardware response includes immediate auditory and haptic feedback through a buzzer and vibration motor, alongside visual warnings via LEDs. Simultaneously, a relay module is triggered to control a DC motor, simulating an automatic braking sequence by reducing vehicle speed autonomously. This integration of high-level machine learning with low-latency embedded hardware provides a robust fail-safe mechanism. Experimental results demonstrate that the system achieves high accuracy in varied lighting conditions, offering a scalable and cost-effective solution for enhancing automotive safety and reducing fatigue-induced fatalities.},
        keywords = {Arduino nano, LED, buzzer, vibration motor, dc motor, relay, buzzer, open cv, scipy, nump, pycharm},
        month = {March},
        }