Coupling Mechanism Analysis And Active Control Method For New Mixed Traffic Flow

Connected and Autonomous Vehicle（CAV）has great potential in improving traffic safety,alleviating road congestion and improving driving comfort,and is even likely to change human traffic mode.It will be the development trend and core of future transportation.Countries all over the world attach great importance to their significant impact on the social economy.CAV is significantly different from traditional Human-driven Vehicle（HDV）in terms of individual information acquisition,perception ability,reaction time,interaction behavior,etc.At the same time,the traffic flow mixed with CAV has some key factors affecting driving behavior,such as the dynamic switch of CAV different driving modes,the interaction between CAV and HDV,and vehicle platooning,which leads to structural changes in traffic flow operation rules.Therefore,for the new mixed traffic flow with a certain proportion of CAV,it is difficult to study its coupling mechanism and active regulation method,which is also the key to influencing the gradual integration of CAV with the existing traffic system.In light of this,the goal of this paper is to increase the operational efficiency of the new mixed traffic flow on high-grade highways.An active traffic flow management and control method is proposed from the vehicle level,lane level,and road network level by analyzing the coupling mechanism of traffic flow under the conditions of different CAV market penetration rates,and the effectiveness of the proposed method is verified using an integrated simulation platform and numerical simulation experiments.Firstly,an analytical expression of the relationship between four types of vehicles,including human-driven vehicles,ACC vehicles,CACC leading vehicles,and CACC following vehicles,is established based on the characteristics of random distribution of vehicle spatial position and vehicle platooning in the new mixed traffic flow.In the analysis of the vehicle coupling mechanism,an analytical method for the stability of mixed traffic flow based on the traditional car-following model structure is proposed,and the basic graph model of traffic flow for lane capacity analysis is constructed by applying the equilibrium"speed-headway"relationship.The mechanism of CAV market penetration rate and maximum platoon size affecting traffic flow stability and capacity is discussed.In the analysis of the lane coupling mechanism,combined with the two-lane road scene,three-lane management strategies suitable for different CAV development stages are proposed,and the traffic capacity of the road roadway under different lane management strategies is analyzed.Secondly,in terms of vehicle-level active control,control methods such as cooperative adaptive cruise control（CACC）,cooperative lane-change control and cooperative merge are proposed.To improve the H₂ performance and H_∞performance of the CACC system,a compound control strategy based on feedback and feedforward is proposed,and a quasi-maximum-minimum feedback controller and a feedforward controller based on generalized extended observer are designed.Considering the coupling characteristics of longitudinal and horizontal motion during the lane-change process,a cooperative lane-changing method including longitudinal spacing adjustment and lane-changing execution is proposed to meet the multi-objective optimization requirements such as lane-changing efficiency,traffic safety and lane-changing comfort.A simulation platform based on Matlab Simulink and Car Sim was built to verify the effectiveness of the cooperative lane-changing method.Aiming at the problem of long delay caused by vehicle deceleration（or stop）in the ramp confluence area,a two-stage model for vehicle trajectory optimization is proposed.The first stage is to optimize the merging sequence of vehicles,and the second stage is to optimize vehicle speed based on the optimal merging sequence.Then,in terms of lane-level active control,an integration method t of adaptive cruise control and variable speed limit is proposed to improve traffic efficiency in bottleneck roadways.Considering the randomness of vehicle distribution and V2V distance limitation,an adaptive cruise control framework including ACC speed regulation mode,ACC/CACC gap regulation mode and hysteresis control mode is proposed.A variable speed limit control model based on the analysis of rear-end collision is established,and the coordination of the two control methods is realized through the variable speed limit value and the desired speed in adaptive cruise control.Simulation results show that the cooperative control method can make full use of the advantages of variable speed limit control and adaptive cruise control to improve traffic efficiency on the premise of ensuring traffic safety.Finally,in terms of network-level active control,a mixed traffic flow optimization method for traffic guidance is proposed.With consideration of the differences between CAV and HDV in environmental information perception and vehicle control,the user equilibrium mode and system optimal mode are selected to characterize the path selection mode of HDV and CAV,respectively.The traffic flow allocation models under normal lane and CAV lane are constructed,respectively.The influence of CAV market penetration,traffic demand and the existence of CAV dedicated lanes on traffic flow is analyzed.The proposed model can provide a theoretical reference for traffic guidance and CAV dedicated lane deployment in the future CAV environment.