2005Unpublished venueRequires access

Monitoring Driver's Alertness Based on the Driving Performance Estimation and the EEG Power Spectrum Analysis

Sheng‐Fu Liang, Chin‐Teng Lin, Ruochan Wu, Y.C. Chen, Tingyu Huang, Tzyy‐Ping Jung

Open publisher page 88 citations

Abstract

Preventing accidents caused by drowsiness behind the steering wheel is highly desirable but requires techniques for continuously estimating driver's abilities of perception, recognition and vehicle control abilities. This paper proposes methods for drowsiness estimation that combine the electroencephalogram (EEG) log subband power spectrum, correlation analysis, principal component analysis, and linear regression models to indirectly estimate driver's drowsiness level in a virtual-reality-based driving simulator. Results show that it is feasible to quantitatively monitor driver's alertness with concurrent changes in driving performance in a realistic driving simulator.

About this research paper

What this paper is about

Preventing accidents caused by drowsiness behind the steering wheel is highly desirable but requires techniques for continuously estimating driver's abilities of perception, recognition and vehicle control abilities. This paper proposes methods for drowsiness estimation that combine the electroencephalogram (EEG) log subband power spectrum, correlation analysis, principal component analysis, and linear regression models to indirectly estimate driver's drowsiness level in a virtual-reality-based driving simulator. Results show that it is feasible to quantitatively monitor driver's alertness with concurrent changes in driving performance in a realistic driving simulator.

Why it matters

OpenAlex reports 88 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Preventing accidents caused by drowsiness behind the steering wheel is highly desirable but requires techniques for continuously estimating driver's abilities of perception, recognition and vehicle control abilities. This paper proposes methods for drowsiness estimation that combine the electroencephalogram (EEG) log subband power spectrum, correlation analysis, principal component analysis, and linear regression models to indirectly estimate driver's drowsiness level in a virtual-reality-based driving simulator. Results show that it is feasible to quantitatively monitor driver's alertness with concurrent changes in driving performance in a realistic driving simulator.

Key concepts: Alertness, Driving simulator, Electroencephalography, Principal component analysis, Computer science, Autoregressive model, Spectral density, Simulation

Related papers

Back to paper searchBrowse research topicsOriginal source
Monitoring Driver's Alertness Based on the Driving Performance Estimation and the EEG Power Spectrum Analysis — Research Paper | ScholarLens