Construction Of Vehicle Lane-changing Model And Analysis Of Traffic Flow Characteristics Of Mixed Traffic Flow

There are great differences in vehicle performance and driving behavior between Connected and Autonomous Vehicles(CAV)and Human-driven Vehicles(HV).It has guiding significance for the control of autonomous vehicles mixed into road traffic flow that not only considering the driving characteristics and interactions between the CAV and the HV but also building a suitable vehicle lane-changing model for human-machine mixed driving traffic flow in various scenes and then analyzing the macroscopic characteristics of traffic flow.The main research work of this paper is as follows:Firstly,the microscopic driving behavior and the vehicle interaction in human-machine mixed traffic flow are analyzed.By considering the differences of operating characteristics between the CAV and the HV,the microscopic driving behavior of the CAV is analyzed.Furthermore,the vehicle following behavior of the HV,the CAV and the CAV’s platoon in human-machine mixed driving traffic flow is analyzed respectively.Considering the different intentions of vehicle,the scenes of lane changing are divided,and the vehicle lane changing decisions are analyzed for different lane changing scenes.Quantifying the psychological effects of CAV on manual vehicle drivers in manner of Driving Behavior Questionnaire,based on that,the changes in driving behavior such as driver’s driving intention to follow,tolerance for slow driving,and extra vehicle spacing etc.are analyzed.Secondly,based on cooperative driving behavior,a vehicle lane changing model for human-machine mixed driving traffic flow is constructed.In addition,parameter calibration and model verification are carried out.Distinguishing passive lane change intention and active lane change intention,this paper calculates the safety clearance of vehicle lane changing under circumstance of different vehicle types in the target lane and analyses conclusions combined with driving behavior.Considering the distance change due to the CAV’s effects on drivers,this paper improves the safe distance model.Three scenes are considered: free lane-changing when the space conditions of the target lane meets the lane change requirements,cooperative lane-changing when space do not meet requirements,and active lane-changing of the CAV to realize formation driving.Combining the changes of driver’s decision-making behavior and vehicle lane-changing benefits the vehicle lanechanging strategy is formulated.Using cellular automata(CA)theory,considering the CAV’s platooning traveling and the CAV’s cooperative lane-changing behaviors,a vehicle lanechanging model of human-machine mixed traffic flow is constructed.Using Matlab to build simulation scenes and simulate the mixed traffic flow,this paper determines the pending parameters which include the CAV’s vehicle lane-changing parameters,maximum platoon etc.and verifies the validity of the built model.Finally,combining the built lane-changing model and the parameter calibration results,the Matlab simulation scene is used to simulate the basic sections of the highway,and the laws of the effects of the CAV mixing to macroscopic characteristics of traffic flow is analyzed from the aspects of traffic efficiency,stability and safety.The results show that the CAV can significantly increase the average speed of traffic flow and the quantity of road flow.When the vehicle density is lower than 40 veh/(km·ln),the CAV can gradually reduce the fluctuation of vehicle speed.However,when the density is higher than 120 veh/(km·ln),the average vehicle speed fluctuation firstly increases and then decreases with the increase of the permeability,the vehicle lane-changing frequency firstly increases and then decreases with the increase of the permeability.And the lane-changing behavior is concentrated in the range of vehicle density of 20-80 veh/(km·ln)and permeability of 0.3-0.8.The CAV improve the road capacity and also affect the safety of traffic flow.Under the condition of low permeability,the risks of vehicle collision increase in fluctuation.When the penetration rate is greater than or equal to 0.8,both the risks of vehicle collision decrease and the safety of traffic flow improve significantly.Based on the interaction of different types of vehicles and the collaborative driving behavior of the CAV,this paper improves the existing model,constructs a vehicle lanechanging model for human-machine mixed traffic flow,and conducts simulation analysis.The model considers vehicle differences to distinguish various lane-changing scenes and considers changes of driver’s behavior intentions,which can accurately describe the driving characteristics of vehicles in a human-machine mixed driving environment,and reflect the impact of the CAV mixing on the macro characteristics of road traffic flow.