Study On The Stress Response Of Drivers Under Different Warning Methods

Active safety warning,which will provide warning information for drivers when a potential danger arises by detecting the surrounding environment of the vehicle,is an effective means to prevent traffic accidents.It is important to select a reasonable warning method,which determines whether the warning information can be transmitted to the driver accurately and timely.At present,overseas and domestic scholars mainly research on the impact of warning methods from the subjective feelings and action responses of drivers.However,there are few results about the drivers' s physiological effects of warning methods,what's more,there are less research results about the optimization of warning signal parameters.Therefore,it is necessary to research on the stress responses of drivers detailedly.This paper first optimizes the parameters of warning signals based on the trend surface model,then analyzes the effects of the physiological changes,subjective evaluation and action responses of drivers under auditory,visual,and tactile warning signals in a single and composite manner,finaly proposes an intelligent warning plan which takes annoyance into account,providing drivers with a more humane warning method.The research results have the important reference significance to design warning methods for car manufacturers.Firstly,the paper analyzes the driver's stress response process under auditory,visual and tactile warnings and selecte evaluation index to evaluate stress response ability,including physiological changes index(the growth rate of heart rate and the growth rate of EEG ?wave power spectral value),subjective evaluation index(urgency and annoyance)and action response index(reaction time).And then designsing simulated driving experiment to collect data of driver's stress response ability evaluation under different warning methods,laiding the foundation for the further study.Secondly,the “warning parameter optimization” experiment was conducted to collect data of the driver's stress response ability under warning signal with different intensity and one time duration.The complex relationship between evaluation index and signal intensity was analyzed to reveal the impact of warning signal parameters on drivers.The auditory,visual and tactile trend surface models were builded to optimal warning signal parameters with the objective of best reaction time.And then,the "single warning" and "comprehensive warning" experiment was conducted to collect drivers' stress response ability under different warning methods.The paper characterize the indicators of action responses,physiological changes and subjective evaluations after studying the significance of difference of different index under different warning methods by variance analysis and non-parametric tests,and analyze the relationship between single warning methods and compound warning methods to provide reference for the design of warning methods.Finally,the mediating effect of driver's physiological indicators and subjective evaluation between response time and warning methods was validated,revealing the process of physiological changes and subjective feelings on the driver 's reaction time;The characteristics of drivers' physiological changes in reaction time are analyzed by using the GA_K-means model for the growth rate of EEG ? wave power spectrum values and reaction time to divide it into phases.The relationship between subjective feelings and reaction time is analyzed to reveal the degree of influence of urgency and annoyance on the driver's response time.As warning methods will have a negative impression,an intelligent warning method that considers annoyance was proposed by using QPSO_RBF neural network to identify driver annoyance.This paper analyzes the impact of warning signal parameters,single warning methods and compound warning methods on drivers' physical,mental and behavioral to reveal the driver's stress response process under different warning methods,providing a reference of the designs of warning signal parameters and warning methods for car manufacturers.