Research On Intelligent Intersection Scheduling Strategy And Vehicle Trajectory Planning Based On Vehicle Road Collaboration

In recent years,with the rapid increase in the number of vehicles on urban roads,traffic congestion has become particularly prominent at road intersections.At present,intersections mainly use signal lights to control vehicle traffic.When the traffic flow is high,signal lights cannot effectively regulate,resulting in traffic congestion.With the development of vehicle networking and autonomous driving technology,intelligent intersection scheduling systems under vehicle road collaboration have become possible,improving the traffic conditions at intersections.The specific research content of this thesis is as follows:Based on vehicle road collaboration technology,an intelligent intersection scheduling system is proposed,which includes intersection scheduling control and vehicle end trajectory planning control.Analyzed the issue of internal conflicts at intersections from a micro level,and established a matrix of corresponding relationships between vehicle conflicts within intersections.The intelligent intersection dispatching system model is built,including intelligent intersection functional area division,intelligent connected vehicle kinematics model,grid resource occupancy and anti-collision model in the core traffic area.Based on the analysis of intersection scenarios,two intelligent scheduling algorithms at the end of the road are proposed: based on the study of the minimum safe headway that vehicles should maintain under different conflict types,a scheduling strategy of first come first served(FCFS)is proposed;In response to the issue that the FCFS policy traffic order is not optimal in many cases,Monte Carlo Tree Search(MCTS)algorithm and some heuristic rules are adopted to solve the optimal traffic order.On the basis of the results of the intelligent scheduling algorithm at the end of the road,vehicle trajectory planning was studied: with vehicle energy consumption as the optimization objective,a vehicle speed planning model in the speed regulation zone was established,and the Krauss mobile following model was used to avoid collisions between vehicles in the same lane in the speed regulation zone;Then,based on dynamic programming and Quadratic programming algorithms,a vehicle obstacle avoidance trajectory planning algorithm in the core traffic area of intelligent intersection is proposed to realize the obstacle avoidance processing of unexpected obstacles in the core traffic area.In order to verify the effectiveness of the proposed roadside FCFS and MCTS scheduling algorithms and vehicle trajectory planning methods,an intersection simulation environment based on Python and SUMO was established.Simulation experiments were conducted on traffic efficiency,environmental and economic benefits,vehicle speed planning in the speed control zone,and obstacle avoidance in the core traffic zone in an intersection simulation environment.The experimental results verified the effectiveness of the roadside scheduling method and vehicle trajectory planning method.