Effects Of Typical Advers Factors On Driving Behavior Performances-A Driving Simulator Based Study

Traffic safety problem as a social issue has received common concerns and it is also an important project in the traffic research area.How to enhance traffic safety and reduce casualties and economic losses caused by traffic crashes has become a vital task and mission for the traffic management researchers from the world.In the complex road system that consists of human,vehicles,roads and environment,drivers are the main decision-makers and controllers of the system,and also the main defenders and executors of traffic safety.In the driving process,effects of adverse environment factors on driving behavior performances are the main causes of traffic crashes.Based on traffic behavioristics,psychology,physiology,cognitive neuroscience and other multidisciplinary theory,this study chose typical adverse factors from outside weather environment,road driving environment and vehicle interior environment respectively as examples,and used the BJTTU high-fidelity driving simulator to study effects of these adverse factors on driving behavior performances and driving safety.The driver’s cognitive process,decision-making process and behavior results as well as their correspondence mechanism were deeply investigated.The generating process of traffic crash and the evolution mechanism between human factor and behavior were further explored.The main research contents contained the following four aspects:(1)Considering fog as the typical adverse weather environment,effects of fog on driver’s behaviors under different risk levels were studied.Three different risk levels combined with three different densities of fog were designed in the experimental driving scenarios of this study.The scenarios could be used to test driver’s basic vehicle-control behavior at low-risk level,the dynamic speed adjusting behavior at mid-risk level,and the emergent speed responsing behavior at high-risk level.The results showed that the drivers were aware of the potential risk induced by fog and tried to reduce speed to compensate for the risk in different scenarios of different risk levels,however,the crash risk in driving still remained high because of the limited visibility resulted by the inclement weather(especially for the heavy fog).In the low-risk scenario,fog weather increased the difficulty of predicting of the abrupt change of road alignment in front,so the driver’s speed control ability and lane control ability were obviously lowered.In the mid-risk car-following scenario,drivers’ average speeds in both acceleration and deceleration stages were lower in fog conditions than in no fog condition.The respond abilitiy of starting acceleration/deceleration action and the control ability of acceleration/deceleration rate were weakened in fog weather.In the collision avoidance scenario of high-risk level,driver’s instantaneous braking speed and collision speed were significantly lower in fog conditions than in no fog condition.When faced with a potential collision,the driver’s deceleration was delayed in fog,which led to that the crash rate increased three times in fog conditions.(2)Considering limited sight distance at non-signalized intersection as the typical adverse road driving environment,effects of sight distance at intersections on the driver’s collision avoidance behaviors were studied.Three grades of sight distance condition were designed at a non-signalized intersection to investigate the changes of drivers’ collision avoidance behaviors,collision avoidance maneuvres and crash risk when encountering a potential right angle collision.The relationship among sight distance,collision avoidance behaviors and crash risk were further investigated.The results showed that more drivers tended to slow down to avoid a collision with the improvement of sight distance.The deceleration actions were started earlier and the deceleration rates were higher as the sight distance increased.Compared to the sight condition of 70m,the crash rates were reduced by 57.3%and 76.5%respectively under the sight distance of 83m and 100m.Besides,it was proved that the traffic conflicts at non-signalized intersections were still inevitable even if the sight distance met the road design requirements.Therefore,it was necessary to encourage expanding sight triangle area to improve the intersection traffic safety in the actual road design or land use planning.(3)Considering cell phone use as the typical adverse vehicle interior environment,effects of cell phone use on driver’s rear-end collision avoidance behaviors were studied.The car-following driving scenario with potential rear-end collision were designed to test the effects of different cell phone use conditions(no phone,hands-free and hand-held)on driver’s rear-end collision avoidance process.The driver’s rear-end collision avoidance mode and the key variables were determined,and the rear-end collision risk assessment model considering cell phone use as interference were established.The rear-end collision risk assessment model established in this study revealed the behavior mechanism in the collision avoidance process and the internal relationship among the key variables.It also evaluated the effects of cell phone use on driver’s avoidance behaviors and collision risk.The results showed that driver’s brake reaction time,deceleration adjusting time and maximum deceleration rate were significantly affected by the use of cell phone,which increased the collision risk.The driver’s avoidance behaviors were also quite different under the conditions of hands-free and hand-held cell phone use.The risk compensation behaviors were more obvious under the hand-held condition,and the crash risk under both conditions had no obvious difference.(4)Taking the rear-end collision avoidance scenario with cell phone use interference as an example,the driver’s eye movements and EEG activities were studied in the process of collision avoidance.According to the data collected by the driving simulation platform,the driver’s eye-movement and EEG characteristics in different stages of rear-end collision avoidance were analyzed,and their correlation analysis with behavior data were conducted.The rear-end crash prediction models based on the independent parameters of eye movements,EEG and driving behaviors as well as the combined parameters were established respectively.With respect to the driver’s eye movements,the results showed that the efficiency of visual information processing was higher for professional drivers than non-professional drivers.Female drivers paid more attention to the sudden deceleration of leading vehicle in mind.The use of cell phone shifted driver’s attention and led to the blink inhibition phenomena.With respect to the driver’s EEG,the power ratio of delta was decreased while the power ratio of beta was increased in the collision avoidance process.The high-frequency waves’ power(alpha and beta)of female drivers were significantly higher than that of male drivers during the normal driving stage and brake reaction stage.The use of cell phone led to the "alpha blocking" in the deceleration adjustment stage.Among the rear-end crash prediction models established by the parameters of eye movements,EEG and driving behaviors,the prediction accuracy of the model with combination of all parameters was highest,which reached up to 97.93%.When the parameters were used independently,the predicition accuracy of the model with EEG parameters was highest(94.3%),and then followed by the driving behaviors parameters(91.8%)and eye-movement parameters(81.9%)respectively.