Research On Traffic Flow Characteristics Of Expressway Weaving Area Mixed With Autonomous Vehicles

With the opening of the "14th Five-Year Plan",the strategy of "transportation power" shoulders an important historical mission,and the construction of urbanization and urban agglomeration will reach a new level.During this period,the fast,safe and green urban transportation network will play a key role.As an important part of the urban road network,expressway undertakes the rapid passage of urban internal traffic and the roundabout passage of external traffic,effectively shortening the travel time.The structure and composition of the expressway weaving area are quite different from the basic road section,which is the bottleneck of the mainline traffic and often causes traffic congestion.The rapid development of intelligent transportation and driverless technology has promoted the use of autonomous vehicles on the road.However,the transition from purely manual to autonomous driving will take a long process,so the traffic flow with both manual and autonomous vehicles will exist on the road for a long time.Mixing with autonomous vehicles,the driving of manual vehicles will be affected,and their car-following and lane-changing characteristics will change.At the same time,the interaction between manual and autonomous vehicles will deeply affect the operation of traffic flow,and the traffic efficiency and driving safety of the traffic system will change significantly.Therefore,it is of theoretical and practical significance to study the traffic operation of expressway weaving area in mixed flow environment.This paper takes the traffic flow mixed with manual and autonomous vehicles as the research object,constructs its car-following and lane-changing model,and analyzes its operation characteristics in the expressway weaving area through numerical simulation.Firstly,the Gipps safety distance and speed are introduced to build the carfollowing model of manual vehicles based on Nasch rules of cellular automata.The dynamics model of PATH Laboratory in Berkeley is used to simulate the car-following behavior of CACC vehicles,and the degradation of CACC into ACC vehicle is also considered.Secondly,an improved cellular automata lane-changing model(ISTCA)is established to solve the problem that the lane change space gap is too conservative and lane change demand is suppressed in the STCA model,considering the influence of vehicles before and after the target lane on lane change.It not only ensures that the vehicle can obtain its desired speed after changing lanes,but also avoids the immediate deceleration that may occur after the vehicle changes lanes and travels for a time step.Finally,a forced lane-changing model is constructed to simulate the weaving operation of vehicles,according to the entering and exiting behavior of vehicles in the weaving area.The simulation results show that when the CACC vehicle penetration rate is less than 0.6,the traffic congestion in the weaving area and its upstream area is severe;when the penetration rate is greater than 0.6,the traffic efficiency in the weaving area is significantly improved,and also the traffic operation.When the length of the weaving area reaches 500 m,the traffic operation of mainline lanes 2 and 3 improves,the traffic congestion index drops significantly,and the mainline flow rate reaches 0.9veh/s.With the rise of the length of the forced lane change area upstream of the weaving area,the flow rate in the weaving area will increase,and the average driving speed will continue to improve.When the weaving flow rate ratio reaches 0.4,the traffic efficiency of the mainline is low,and the traffic safety hidden danger in the weaving area is high.