Risk Perception Prediction Method For Lane Change Behavior

The transportation system consists of "people-vehicle-road-environment".In modern transportation,vehicles mainly refer to motor vehicles.Safe driving behavior plays a crucial role in the safe and orderly operation of transportation systems.The factors affecting driving behavior include the driver’s own conditions as well as the external traffic environment.Their own conditions include the driver’s physical health status,and the external traffic environment includes traffic flow saturation and traffic control conditions.Heart rate is not easily affected by other factors during the collection process,and this indicator mainly reflects the activity of body functions.Heart rate can be used as an indicator to evaluate the health level of the driver.Eye movement refers to the change of the driver’s eye trajectory.It is used to evaluate the driver’s level of attention to the external traffic environment,and to a certain extent can reflect whether the driver is fatigued.The vehicle behavior is the result of the driver’s operation output.This article uses the vehicle speed as the evaluation index,which can reflect the current road service level.This article uses heart rate,eye movement,and vehicle speed as indicators for evaluating driving behavior.In order to obtain a safe level of lane change behavior,a driver risk perception quantification value is proposed to evaluate the driver’s ability to identify road risks.Hidden Markov model is used to predict the quantitative value of driver risk perception in combination with heart rate level,eye movement behavior and vehicle behavior.The paper first elaborates the analysis methods of heart rate,eye movement behavior and vehicle behavior,and determines the research indicators.At the same time,the data acquisition test program is introduced,including the use of test instruments,test path selection and so on.Then analyze the driving behavior indicators under different influencing factors,and use the one-way analysis of variance to obtain the significance of driving duration and road service level on driving behavior.Then proposed the driver’s perception of the current road environment risk quantification,determined objectively and subjectively.Objectively refers to the real-time video processing according to the later period,and the judgment value of the environmental risk;Subjectively refers to the subjective feelings of the driver during driving.The ratio between the objective and subjective risk quantification value is the driver’s own risk perception quantification value.The risk perception quantification value ≤ 1 indicates that the driving behavior is safe,and the >1 indicates that the driving behavior is at risk and needs to be given a safety warning.Because the driving behavior index is observable,the driver risk perception quantification value is difficult to obtain in the same way compared to the driving behavior index.Based on this characteristic,the Hidden Markov Model(HMM)is applied to the risk of the driver.Perceived quantified values to predict.The driving behavior parameters under different risk perception quantification values are analyzed to obtain the level of the observed sequence of risk perception quantification values.The results show that the model can better predict the driver’s risk perception quantification value and improve road safety.