Study On The Analysis And Simulation Of Phantom Traffic Jams Considering Autonomous Driving

With the rapid increase in car ownership,traffic jam has become a growing problem that can’t be neglected.Causes leading to traffic jam generally include traffic accidents and too many road intersections and so on.However,there always emerges a type of ‘phantom’ traffic jam where you are not able to find the apparent causes for it.The moving traffic suddenly slows down as a whole and after a while it suddenly becomes smooth again.This kind of phenomenon is being called ‘Phantom Traffic Jams’.Although you can’t find the superficial causes for it,however,it finally comes down to the different reactions of drivers to the slowdown or speedup of the cars in front of them.This research paper aims to explore the impact of automatic driving on ‘phantom traffic jams’ by employing the research method of using cellular automata model.Based on previous researches on ‘phantom traffic jams’ and the hybrid traffic flow of human driving and automatic driving,and in view of the differentiating behavioral characteristics of automatic driving from human driving,this paper also explores the different control tactics that may be employed by automatic driving vehicles under different scenarios.Finally,analysis and research are carried out into the impact of automatic driving on ‘phantom traffic jams’.Detailed content in the paper includes:First,this paper gives a brief introduction to the development of automatic driving and current researches about it.It also summarizes the research content on ‘phantom traffic jams’ and the modeling foundations of cellular automata in automatic driving and the hybrid traffic flow of automatic driving and human driving.Second,on the basis of the previous researches,this paper places the focus of the analysis into the causes of ‘phantom traffic jams’ phenomenon on the control tactics of autonomous vehicles.In view of the fact that autonomous vehicles can capture information about car locations and the speed of the cars in front and at back,this paper proposes the CA model for automatic driving based on safety distance.This model takes into consideration the impact of the cars in front and at back on the cars that follow them and proposes different driving control tactics for different driving scenarios.Meanwhile,on the basis of the existing CA model for human driving,the CA model for the hybrid traffic flow of human driving and machine driving is also established.Then,this paper uses MATLAB simulation software to simulate the human driving model and the automatic driving model.The analysis focuses on the proportion of autonomous driving and the reaction time of autonomous driving,and whether to consider the influence of autonomous driving on improving ‘phantom traffic jam’ phenomenon in the case of rear car.The research findings indicate: the advantages of the car-following control tactics for automatic driving compared with those for human driving lie in the fact that it can adjust its own speed so as to lower the speed fluctuations of the entire traffic flow and reduce frequent stopping and restarting,thus improving ‘phantom traffic jams’.The introduction of autonomous vehicles into the hybrid traffic flows plays a remarkable role in lifting the speed of the road traffic,improving road capacity and easing ‘phantom traffic jams’.Finally,in order to verify the deterministic deceleration of autonomous driving and its convergent driving style is the mechanism that can improve the ghost congestion.From the cause of the ‘phantom traffic jam’ phenomenon,this paper discusses the variation of the probability of occurrence of different control strategy scenes with the vehicle density and the proportion of auto-driving on the road,and analyzes the interaction between manned and autonomous vehicles in mixed traffic flow.