Research On The Risk Of Large Angle Lane Changing Of Vehicle Under Congested Traffic Based On Driver’s Physiological Characteristics

The imbalance between the supply of transportation infrastructure and road demand leads to more serious urban road traffic congestion.Traffic congestion has a great impact on the driver’s psychological state and driving behavior,and even induces traffic accidents.Lane changing behavior is a basic driving state of vehicles.Under high-density traffic flow conditions,lane changing operation is more difficult,causing more traffic conflicts and greater driving risk factors.Therefore,from the perspective of human factors engineering,this paper associates the driver’s visual,physiological and psychological state with the risk of lane changing,and provides reference and theoretical basis for identifying road traffic risk points and reducing traffic accidents.In this paper,the large angle lane changing behavior is defined,and five typical large angle lane changing scenarios are selected,including short-distance continuous lane change,using the contraflow left-turn lane,parallel turning,bottleneck confluence,and turning around.The urban road congestion traffic simulation scene is built,and the driving simulation test is carried out to obtain the visual,physiological and driving data of the driver under each lane change situation.Based on the changes of visual,physiological and driving indicators,the driver’s mental load and road traffic risk are judged,and the risk factors leading to driver tension in the environment are analyzed.The entropy weight-TOPSIS comprehensive evaluation model is established,and the five large angle lane changing behaviors are divided into 13 evaluation objects in different stages,and the risk degree of large angle lane changing behavior is evaluated.Through the questionnaire survey,the driver’s subjective feeling of each large angle lane changing behavior and the simulation degree evaluation of the driving simulation test are obtained.The results show that visual indicators such as continuous fixation time,fixation interest area,cumulative saccade time,saccade amplitude,pupil area change rate and physiological indicators such as ECG and EDA can better characterize the driver’s psychological changes in the case of large angle lane changing.Based on the change law of visual and physiological indexes,the risk degree of the above five large angle lane change behaviors is sorted,which is using the contraflow left-turn lane>continuous lane changing>rotary confluence>bottleneck lane changing>parallel turning,and the mean change rates of pupil area were 26.28%,24.22%,21.40%,17.74%and 14.83%,respectively.By using the pupil area change rate,heart rate growth rate and skin electricity growth rate,the high-risk points of the five large angle lane change processes are the S-turn at the opening of the central separation zone,driving into the second target lane,the head enters the opposite lane,changing lane to open road,and the central area of intersection,and the potential danger of the position is analyzed.The entropy weight-TOPSIS comprehensive evaluation model shows that the relative proximity C_i of the above high-risk points to the optimal solution was 0.937,0.813,0.688,0.495 and 0.440,respectively.And the safety risk of the whole process of left turn and continuous lane changing is large,and the risk degree of the lane changing execution stage of each lane change behavior is the largest.The results of the questionnaire show that the proportion of people who feel very nervous in using the contraflow left-turn lane and continuous lane changing is the highest,and the simulation effect of the driving simulation test is better.