Conflict Analysis Of CAVs At Unsignalized Intersection Based On Cooperative Vehicle Infrastructure Control

With the development of automatic driving and Internet of Vehicles,more and more attention has been paid to the Cooperative Vehicle Intersection Control at unsignalized intersections.This paper assumes that the intelligent roadside infrastructure has been laid near and inside the unsignalized intersection,and the communication with Vehicles is in good condition without packet loss,Connected and Automated Vehicles（CAVs）,As the main driving body,CAVs are used to analyze the conflict of Vehicles in automatic driving mode when they pass through unsignalized intersections in the Cooperative Vehicle Infrastructure environment,The specific research work of this paper includes:（1）A conflict detection method of CAVs based on rectangular model is established.The rectangular edge and endpoint,edge and edge,and intersection of endpoint and endpoint are regarded as risk indicators.The coordinate range of vehicle conflict location is determined,and under the condition that DSRC communication protocol is used between vehicles,the braking distance model of automatic driving vehicle is established by replacing the driver’s reaction time with the system reaction time,and the rectangular conflict detection model is improved as the safety distance.（2）Based on the idea of intersection reservation control,according to the relevant road safety regulations,the traffic Library of unsignalized intersection is established.When CAVs and HDVs are mixed,the digital access library is used to allocate the right of way,so as to avoid vehicle conflict from the traffic sequence,and the unsignalized intersection Access Library under Cooperative Vehicle Intersection environment is initially established.According to the research direction of trajectory optimization,a conflict resolution algorithm based on rectangular conflict detection model is proposed to guide the speed of Vehicles without right of way and avoid the intersection of vehicle space-time trajectory in the conflict area.（3）Based on the open source framework of the Internet of vehicles,sumo and OMNe T++are coupled to each other.Through the verification of the conflict detection model and resolution model,compared with the simulation results,it can be seen that the prediction success rate of the conflict detection model is very high,and the values taken from the resolution speed range can also effectively avoid the conflict between the two vehicles.（4）Compared with the traditional unsignalized intersection control method,the vehicle road collaborative speed guidance method has better performance in terms of average vehicle speed,total driving time and vehicle CO₂ emission than the traditional deceleration to parking resolution method.Finally,the Python tool is used to analyze the overall running state.By comparing and analyzing the differences between the two control methods in"mean Halting Duration","mean Max Jam Length In Meters","mean Speed"and"mean Time Loss",it shows the effectiveness and feasibility of the proposed algorithm.