EEG-based analysis of human driving performance in turning left and right using Hopfield neural network

Abstract

In this article a quantitative analysis was devised assessing driver’s cognition responses by exploring the neurobiological information underlying electroencephalographic (EEG) brain signals in a left and right turning experiment on simulator environment. Driving brain signals have been collected by a 19-channel electroencephalogram recording system. The driving pathway has been selected with no obstacles, a set of indicators are used to inform the subjects when they had to turn left or right by means of keyboard left and right arrows. Subsequently in order to remove artifacts, preprocessing is performed on data to achieve high accuracy. Features of signals are extracted by using Fast Fourier Transform (FFT). Absolute power of FFT is used as a basic feature. Scalar Feature selection method is applied to reduce feature dimension. Thereafter dimension-reduced features are fed to Hopfield Neural Network (HNN) recognizing different brain potentials stimulated by turning to left and right. The performances of HNN are evaluated by considering five conditions; before feature extraction, after feature extraction, before reduction of features, after analyzing reduced features and finally subject-wise Hopfield performances respectively. An increase occurred in each level and continued until it has reached its highest 97.6% of accuracy on last condition.

Change in Heart Rate Variability Indexes due to high driving workload in turning left at the intersection in real road environment

“…In order to meet the test requirements, subjects were required driving on a known route in urban road using a same car. Mitra Taghizadeh-Sarabi [8] introduced a new procedure analyzing turning left and right during driving with constant speed in a pre-designed path. In the experiment, we refer to the above driving scenarios, and varied automotive steering as the independent variable we assumed this will lead to a higher driver workload.…”

New Insights on Learning Rules for Hopfield Networks: Memory and Objective Function Minimisation

“…From purely engineering standpoint, in the recent literature Hopfield neural networks have been applied for image processing [20], [9], solving various combinatorial problems [32] [25], [18], random numbers generation [30], [7], and have even been used in conjunction with Brain Computer Interfaces [12], [28]. To increase the energy efficiency of the implementation of Hopfield Networks in hardware, learning in sparse networks have been considered in [29] and subsequently tested in the context of image restoration.…”

Driving Style Recognition Based on Electroencephalography Data From a Simulated Driving Experiment

“…A number of studies have utilized electroencephalography (EEG) to identify dangerous driving states, such as fatigue and distraction (Chuang et al, 2015; Hajinoroozi et al, 2016; Belakhdar et al, 2018; Guo et al, 2018; Ma et al, 2018), driving behaviors, such as emergency braking (Haufe et al, 2011), speeding (Lutz et al, 2008) and turning (Taghizadeh-Sarabi et al, 2013), and driving styles, such as car-following and obstacle-dodging (Lin et al, 2006b; Yang et al, 2018). Specifically, some researchers classify and assess the driver's behavior and style based on the amplitude and power spectral density information of α, β, δ, and θ bands of EEG signals.…”

“…For example, Lin et al (2006b) used the power spectrum analysis to investigate the correlation between driving style and brain activities revealed by EEG, and found power difference at 10 Hz and 20 Hz between aggressive and conservative drivers. Taghizadeh-Sarabi et al (2013) extracted the absolute power of these four bands by Fast Fourier Transforms (FFT) to assess the driver's cognitive responses when turning left and right. Yang et al (2018) combined the amplitude and the power spectral density to classify the driver's driving skill and driving style.…”