Enhancing Driver Safety Through EEG-Based Drowsiness Detection: A Hybrid Model Using Fuzzy Logic and XGBoost

View at publisher Summarise with VIRAT
Authors: Keerthi A. V. Kumar, Vandana Reddy, Debarka Mukhopadhyay
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Driver fatigue is one of the significant causes of traffic accidents; therefore, real-time detection of driver fatigue is crucial. This research gives a configuration that uses fuzzy logic and XGBoost, developed in Python with the scikit-learn library, to identify fatigue from electrical brain signaling (EEG). The calculation of uncertainties is facilitated by fuzzy logic, while the XGBoost algorithm of analysis is both accurate and speedy. The efficiency of this hybrid model was subsequently contrasted with traditional machine learning algorithms such as the Support Vector Machine (SVM), Random Forest (RF), and Gradient Boosting Classifier algorithms, all of which were developed using Scikit-learn. The results prove that the hybrid model is more effective as it tends to yield more accurate results with highest accuracy of 83.40 % and take shorter time to detect than the other models with fastest inference time of 0.193 seconds. This approach points out the importance of integrating a real-time drowsiness detection system by preventing the high number of accidents resulting from fatigue. The research worked with a pre-processed EEG dataset from Kaggle, and all the modeling and testing have been performed in the Google Colab environment for the convenience of the development.

Fields of Study & Subjects

Sleep and Work-Related Fatigue EEG and Brain-Computer Interfaces Human-Automation Interaction and Safety

Publication details

Journal / source
CRC Press eBooks
Pages
156–168
Publication date
26 March 2026
Language
EN
Document type
Book chapter