Research On Mixed Intersection Traffic Strategy Based On Platooning And Dynamic Priority Under The Background Of Vehicle-road Collaboratio

With the increasing maturity of autonomous driving technology,vehicles with autonomous/assisted driving functions are gradually increasing on urban roads.In the foreseeable future,autonomous driving technology will tend to mature and enter the market on a large scale,and autonomous vehicles will become an important part of urban transportation.For a long time,there will be a coexistence of autonomous and humandriven vehicles at intersections,which will lead to a series of problems.For example,different types of vehicles have different driving methods and reaction time.However,the current commonly used traditional traffic management strategy regards autonomous and manual driving vehicles as the same unit for command and dispatch.This fails to make use of the characteristic of autonomous driving vehicles as smart terminals,resulting in safety hazards at intersections and lower traffic efficiency.Therefore,it is necessary for us to proactively design a new traffic management strategy and verify its feasibility in future scenarios.Based on such needs,the research contents of this thesis are as follows:(1)Multi-vehicle dispatching traffic flow control strategy based on platoon formation and dynamic priorityIn a mixed intersection scenario,vehicle groups are established based on the distance between vehicles.Different passing logics and priority calculation methods are formulated for different types of leader groups,and the information of the groups and priorities is updated in real-time,according to the vehicle’s driving status.This effectively prevents autonomous driving vehicles from decelerating to form groups and ensures the fairness of intersection passing.Meanwhile,a strategy based on pure tracking control and cooperative adaptive cruise control is proposed to solve the motion planning for followers,which avoids the high computing power required by all vehicles to use model predictive control.The experimental results in the SUMO traffic flow simulation software show that this passing strategy is better than the traditional strategy in terms of traffic capacity,average speed,average waiting time,and average fuel consumption.(2)Construction of vehicle-road cooperative platform and experimental verificationThe self-built vehicle-road cooperative platform uses Open CV to obtain real-time motion states of all unmanned vehicles under multiple cameras and makes motion planning for the vehicles based on the algorithm mentioned in this thesis.The motion planning is sent to the vehicles through the Zig Bee module.The vehicles receive the upper computer’s instructions,analyze and process them,and control the motor and steering gear to achieve real-time motion control of the vehicles and track the preset route.Based on this platform,experiments are designed to verify the conclusions of the mixed intersection traffic strategy generated on the SUMO simulation platform.