Vehicle-Intersection Coordination For Mix Traffic Flow In An Intelligent And Connected Environment

Facing the condition of mixed traffic including connected and autonomous vehicles(CAVs)and human-driven vehicles(HDVs),the inherent "stop-wait-go" mode of traditional signal control seriously weakens the technical advantages of CAVs for information interaction and coordinated operation.Thus,what kind of traffic mechanism and control method should be adopted at intersections is a critical problem to be solved urgently in the discipline of traffic engineering.Based on the real-world application demand,this paper combined the trends of urban transportation in China and considered the impact of new technology simultaneously.This study summarized the issues in intersection management research and markets and sorted them out as a series of research gaps to fill in the vehicle-intersection coordination designing and modeling for mixed traffic flow in an intelligent and connected environment.The vehicle-intersection coordination model for mixed traffic flow in an intelligent and connected environment in this study is fully extracted from real-world issues with the research goal of realizing the optimization models in solving different modes in specific situations.This study can fulfill some blanks in the intersection management mode to suit different development stages of intelligent connected,and provide systematic theoretical support for solving urban intersection congestion.In this study,autonomous intersection management models are proposed in various backgrounds which covered building an intersection cooperative control mechanism,exploring resource allocation models that take into account both CAV and HDV traffic requirements,reasonable use of CAV market penetration to improve traffic efficiency,and quick accident congestion mitigation.The main work and research results of this study are summarized as follows:(1)This study goes through the related studies to vehicle coordination in an intelligent and connected environment and discusses the development status of intelligent connected technology,modeling and solution of intersection collaborative control and accident congestion mitigation,and traffic simulation methods.Then this study proposed the research objectives,contents,and methods of vehicle coordination for mixed traffic flow in an intelligent and connected environment.(2)This paper analyzes and summarizes the issues involved in the management of intelligent connected intersections,and designs the framework of the intersection coordination method from the traffic control layer and the vehicle control layer.To test the model proposed in this paper,a simulation platform based on a parallel road network system is proposed.(3)To deal with the traditional signal control modes that face the conflicts of unsuitable with the development of intelligent connected technology,this study deal with this issue by a reservation-based intersection mechanism(RIM)method.By designing a unified expression of the spatial-temporal supply and traffic demand at intersections,the resource allocation is computable.On this basis,an intersection spatiotemporal grid resource allocation model is established.Finally,an intelligent decision support model for signal control is proposed.The model assigns green phases based on queue lengths and vehicle waiting times in each lane.Experiments show that the proposed method outperforms the fixed signal control and actuated signal control methods in the condition without CAV participation,and is more suitable for dynamic traffic flow with changing steering proportion.(4)To deal with the lack of resource allocation models that take into account the traffic demands of CAVs and HDVs,this paper proposed a RIM-based vehicle-intersection coordination method with a CAV-dedicated lane.First,lane functions of the reservationbased intersection control method are re-divided,then a CAV lane change model is introduced to support vehicle formation,and finally,the CAV trajectory is optimized.In the solution process,this paper solves the multi-objective nonlinear problem through the PSO algorithm,which effectively improves the solving speed.In this model,the traffic capacity of the intersection increases with the CAV market penetration rate.When the CAV penetration rate is greater than 45%,it is superior to the fixed signal control and actuated signal control in terms of traffic capacity and delay.(5)To deal with the lack of intersection collaboration models that suit any CAV penetration scenario,this paper proposed a RIM-based vehicle-intersection coordination method without CAV-dedicated lanes.First,lane functions of the vehicle-intersection coordination method with a CAV-dedicated lane are re-divided to make it suitable for low CAV penetration scenarios.Finally,an ecological driving model that takes into account fuel consumption,emissions,and delays is constructed.In this model,even if the penetration rate of CAV is zero,it outperforms the fixed signal control and actuated signal control methods in terms of delay,fuel consumption,and emissions.As the CAV penetration rate increases,the performance of the intersection improves steadily.When the CAV penetration rate reaches 100,the vehicle delay at the intersection is almost eliminated.(6)To deal with the rescue and congestion mitigation of intersection traffic accidents in an intelligent and connected environment,this study proposed a vehicle trajectories control method after intersection traffic accidents.The method focuses on congestion mitigation and rescue at the accident site.First,by locating the accident location in the intersection,redistributing the vehicle trajectories,and formulating a new vehicle coordination model,the vehicle avoids obstacles and achieves coordination simultaneously.Finally,the rescue time is shortened by setting up a special rescue lane.In addition,the model also considers the implementation of vehicle guidance at the level of the mesoscopic traffic model,so that the vehicle can effectively avoid the accident intersection and reduce the traffic demand of the accident site.This model provides theoretical support for mitigating intersection accident congestion and improving the safety of the intersection in an intelligent connected environment.This study meets the needs of the intelligent connected development tendency in China,with a background of intelligence,automation,and electrification,which bring great opportunities in transportation.This study mainly focused on vehicle-intersection coordination for mixed traffic flow in an intelligent and connected environment and did research on vehicle coordination methods,resource allocation model,CAV penetration level,accident congestion mitigation,etc.The research results provide theoretical support for intersection management and control,which is of great significance for promoting intelligent connected and driverless technologies in the future.