Traffic Modeling And Simulation Of Safe Driving Behavior

The paper focuses on the study of one-dimension CA model for single lanetraffic flow. Starting from the research of the mechanism of driving behavior, weanalyze serial complex psychological activities of the driver during the drivingprocess in the road and get the main factors of affecting the driver’s control andguidance behaviors. Moreover, such factors are abstracted as some evolution rules ofthe improved CA model, and the numerical simulations are conducted, which verifythe rationality of the improved CA model. The main results achieved in thedissertation can be summed up as follows:First of all, the evolution rules of the one-dimension CA model for single lanetraffic flow are analyzed, and three main decision points are obtained from themechanism of driving behavior in the real traffic, which are (a) the relative velocitybetween the two successive vehicles,(b) the gap between the two successive vehicles,and (c) the driving behaviors determined by the results of decision points (a) and (b).Moreover, the driving behavior is expressed by the randomization probability, andsome complex traffic phenomenon is reproduced by the numerical simulations. In theend, we have known that the relative velocity and security gap have an importanteffect on the driving behaviors and the final traffic state by contrasting the numericalsimulations.Secondly, the effect of the following vehicle’s horn stimulation on the driver’spsychological activities and driving behaviors is studied for single lane traffic flow(overtaking isn’t considered). The influence of ranging condition of the followingvehicle’s horn、the driver’s horn stimulation threshold and sensitivity threshold on thedriving behaviors is analyzed in detail. Besides, the driving behaviors are expressedby the improved acceleration and braking rules in the CA model. It’s obviouslyknown that the following vehicle’s horn stimulation indeed produces an significanteffect on the driving behaviors and the final traffic state by numerical simulations.Then, through contrasting different CA models’ performance we can get that the CAmodel considering the effect of the following vehicle fully has the best performance.Therefore, the following vehicle’s potential influence shouldn’t be neglected in studyof the CA traffic flow models. Thirdly, in the CA traffic flow model which considers the following vehicle’shorn stimulation, the three main factors of S->J phase transition in the heavysynchronized flow are analyzed, which are the ranging condition of the followingvehicle’s horn、the driver’s horn stimulation threshold and sensitivity threshold, andthe mechanism of S->J phase transition are revealed, which has an important guidingsignificance when explaining the cause of traffic jams in the real traffic. In the end,S->J phase transition is observed by numerical simulations, which verifies therationality of the above analysis.