01697367

Component

A vigilance detection method for high-speed rail (HSR) drivers based on Convolutional neural networks (CNN) and wireless wearable electroencephalographic (EEG) collection technology is proposed in this paper. The proposed method consists of three parts: wireless wearable brain-computer interface (BCI) system for EEG signal collection, preprocessing of EEG data and HSR driver vigilance detection. First, a wireless wearable BCI system with eight channels is designed to collect the EEG signal of the HSR driver. Next, the EEG signal is preprocessed with a wavelet de-noising algorithm and down-sampled to enhance the quality of the EEG data and divided into training set and testing set. Furthermore, using the convolutional neural network to train through the training set, and the testing set inputs the trained network to learn and estimate the HSR driver vigilance level. The experimental results show that the method presented in this paper shows excellent performance

This paper was sponsored by TRB committee AR070 Standing Committee on Railroad Operational Safety.

19-04060

English

16p

2019

Transportation Research Board 98th Annual Meeting

Digital/other

Figures; Maps; Photos; References (33) ; Tables

Driver monitoring; Heart rate; High speed rail; Locomotive engineers; Physiology; Technological innovations; Vigilance

Operations and Traffic Management; Railroads; Safety and Human Factors

Transportation Research Board Annual Meeting 2019 Paper #19-04060

TRIS, TRB, ATRI

Dec 7 2018 9:25AM