Modeling Of Selection Behavior On Exit Lanes And Settings Of Guideline Markings For Left-turning Vehicles

With continuous increase of car ownership in China,people’s travel demand is rising.Congestion and accidents have become the focus of all sectors in society on urban roads.For this reason,traffic engineers have taken various measures from many aspects.However,as one of factors causing traffic congestion and accidents,the traffic conflict among different vehicles in the same direction traffic flow at intersections is the most easily ignored factor at present.How to guide co-directional vehicles to pass through intersections safely and efficiently has become an important issue.Based on the grey theory,this paper establishes the traffic volume prediction model.According to the random utility theory,the exit lane selection model is established.According to the left-turning vehicles trajectory characteristics,three models describing the vehicle turning trajectory are established.Considering the layout of intersection,the exit lane selection behavior and the trajectory characteristics of left-turning vehicles,the vehicle’s driving trajectory are proposed.The left-turning guideline markings design method is proposed and a series of simulation models are established.In order to systematically expound the above models and methods,this paper is divided into five parts: The first part systematically analyzes and summarizes the research background and significance,present researches at home and abroad,objectives and contents,and methods and framework.The second part elaborates on the details.The grey theory is combined with trigonometric function and Markov to apply traffic volume prediction.The validity of the model is verified by a practical example.The third part analyzes the selection characteristics of the left-turn vehicle exit lane,and applies Multiple Logit models to turn left.The vehicle model,vehicle speed,turning angle,front vehicle model,different with the preceding vehicle speed,and the choice of the front lane are variables,and the exit lane selection model is established to realize the prediction of the left-wheel vehicle exit lane distribution result.The fourth part analyzes the left-turn vehicle.Based on the various characteristics of the turning trajectory,the circular curve,flat curve and function curve model of the vehicle trajectory are respectively established,and the model is calibrated and evaluated.The fifth part is designed with a circular curves and a plane curves for guideline markings,and the simulation model is established to analyze the guideline type and number to affect the left-turning traffic flow.The results show that when the grey model is used to predict the sequential traffic flow rate,adjusting the traffic flow sequence with cot(x)function can significantly improve the prediction accuracy;the choice of lane entry,the type of vehicle itself,the type of vehicle in front of the same lane,the choice of lane in front of the same lane and the intersection radian of the same lane affect the choice behavior of vehicle exit lane,and the lane efficiency constructed with this as the variable function can be used to predict the distribution of vehicles in exit lanes;function curve model is suitable for describing the trajectory of left-turning vehicles at relatively regular intersections;the trajectory models with circular and plane curves can be used to describe the trajectory of left-turning vehicles at regular or irregular intersections;Left-turning traffic flow guided by plane curves can pass through intersections more efficiently.