Investigation On Traffic Efficiency Of Expressway Access With Intelligent & Connected Environment

With the progress of urbanization,traffic congestion is becoming more and more common,and ground transportation has become an "unreliable" way of travel.Urban expressway,as the main channel of middle and long distance and the link between urban groups,is often congested at peak times,impacting on the urban residential trip efficiency.As the mainstream direction of future traffic development,IOV(Internet of Vehicles)is a new idea to alleviate traffic congestion.First,this study analyzes the mechanism of urban expressway access becoming traffic bottleneck,and introduces two classical control strategies of collaborative merging and two classical control strategies of mainline variable speed limit.Second,aimed at the improvement of the traffic efficiency of the on-ramp,this study puts forward the applicable conditions of the model.Based on analysing the generation of virtual platoon and the decision of cooperative lane-changing,an improved cooperative merging control model with virtual platoon is constructed,in which vehicle control in the inner lane of the main line is included.Third,aimed at the improvement of the traffic efficiency of the off-ramp,the applicable condition of the algorithm is proposed.On the basis of combing the Markov decisionmaking process(MDP),a mainline variable speed limit control algorithm based on reinforcement learning is proposed,And the state space,action space and reward function of Q-Learning algorithm are constructed,and the relevant model parameters are calibrated.Finally,according to the COM interface of VISSIM and Python,the simulation platform is built,the experimental schemes of cooperative merging and the main line variable speed limit are designed.The simulation results show that the proposed improved cooperative merging model and the classical control can improve the traffic efficiency of the entrance area compared with the case without control,And the model studied in this study is improved on the basis of the classical model;Compared with the case without control,the variable speed limit control of the main line based on reinforcement learning and the variable speed limit control of the main line based on feedback can improve the exit area traffic efficiency,and the algorithm used in this study is better than the algorithm based on feedback.