Research On The Influence Of Emotions On Driving Behaviors

With the growth of motor vehicle use in China,the amount of casualty and property loss is ever increasing.According to the annual statistical report of road traffic accidents in China,the number of traffic accidents,fatality,and injury have been declined.However,as the data revealed,the number of fatality and injury per accident was increasing.The emotion is still one of the main causes of traffic accident.Therefore,it is important to analyze the influence of drivers' emotions on driving behaviors and to warn them the dangerous behaviors for traffic safety.In this context,the article clearly raises problems of experimental design,driver emotion recognition,the influence of different emotions on behavior and the prediction of driving behavior as focal point in this research.Secondly,experiments were designed to collect physiological signals and driving behavior data of subjects.Trigger the drivers' four basic emotions of happiness,anger,sadness and fear by different driving simulation scenarios,and collect the drivers' physiological data and driving behavior data in different emotions through the simulated driving platform.Then,the physiological data were pre processing and feature extraction,using SVM(support vector machine)based on binary tree to set up emotion model.And then,recognition the driver's emotion.On the basis of emotion recognition,the driving behavior characteristics under the state of joy,sadness,fear and anger were analyzed Then,a three-level causal decomposition(arousal—reaction--vehicle state)is proposed to further analyze the influence of emotional valence and arousal on driving behaviors.By using the baseline signal elimination method,the change value of the driver cardiac RR intervals,absolute steering and the maximum lane departures under different emotional sources are analyzed.Finally,according to the mathematical description of the problem,a driver model integrating driver's emotion changes was established,and the neural network architecture is designed based on the model.By using the competitive learning algorithm,the drivers' behaviors are predicted.And the driving behaviors under the effct of different emotions were predicted,and the predicted value is very close to the observed value.In the case of high arousal emotional,the average speed increased significantly compared with that in the case of calm,accompanied by obvious overspeed behaviors.In the case of low valence and high arousal emotional,accidents are more likely indicating more dangerous driving behaviors.In the case of low valence and high arousal emotional,the average speed decreased significantly compared with the calm state,indicating more cautious driving behaviors.