Research On Collaborative Control Method Of Reverse Variable Lane Intersections Under Mixed Driving Environment

Intersection are the key nodes in the road traffic network,and it is the main place where traffic flows converge and disperse,as well as the main place where traffic jams occur.In order to improve the efficiency of traffic and relieve traffic congestion at intersections,domestic and foreign researchers have proposed a series of design and control methods of unconventional intersections,and put them into practice.Reverse variable lane intersection is a new method of unconventional intersection in recent years,and it has been applied in many cities in China.It mainly uses the exit lane dynamically according to the signal cycle to improve the capacity of left-turn vehicles,reduce the average delay.Related studies show that the reverse variable lane design not only improves the traffic efficiency of the intersection,but also has safety problems.Its stability is poor,and it is easily affected by microscopic driving behavior.Automated driving technology and Internet of Vehicles technology are considered to be effective methods to relieve traffic congestion and improve traffic efficiency of the intersection.For a long time to come,there will be both connected automated vehicles and human pilot vehicles on the road.Therefore,following the academic frontier,this paper studied the setting method of reverse variable lane and the collaborative optimization method of the intersection under mixed driving environment of connected automated vehicles and human pilot vehicles.The main research contents of this paper include:(1)The traffic characteristics of reverse variable lane intersections are analyzed.Firstly,the operation process of reverse variable lane is described,and the traffic characteristics of reverse variable lane intersections under traditional human driving environment are analyzed based on survey data.Secondly,the influence of connected automated vehicles on traffic flow is analyzed theoretically with the following model.Then,through simulation and theoretical analysis,the traffic characteristics of reverse variable lane intersections under mixed driving environment are studied,and the discrete model of heterogeneous traffic flow platoon and the vehicle following model are established based on mixed driving environment.(2)The setting method of reverse variable lane intersections is analyzed under mixed driving environment.Firstly,the control principle of reverse variable lane intersections under mixed driving environment is expounded.Secondly,the applicable conditions of reverse variable lane design are put forward under mixed driving environment,including basic setting conditions,geometric conditions,traffic volumes,intersection spacing,and the phase sequence of reverse variable lane intersections is analyzed.Then,combining with the road traffic sign marking standard and the actual intersections setting,the design method of reverse variable lane supporting facilities is put forward under mixed driving environment.(3)A collaborative optimization model of speed guidance and signal timing at reverse-variable lane intersection is built based on mixed driving environment.By analyzing the relationship between speed guidance and signal timing,a cooperative optimization model of speed guidance and signal timing is established based on double-level programming.In the lower programming,the vehicle speed guidance model based on MPC is established to minimize vehicle fuel consumption.In order to minimize the average delay,an optimization model of signal timing is established based on double-loop control structure.The solution methods of upper and lower models are analyzed,and the IGA algorithm and the SFLA algorithm are proposed to solve the upper and lower models respectively.At the same time,the rolling optimization strategy of the collaborative optimization model of reverse variable lane intersection is analyzed,and the overall collaborative optimization framework of intersection is proposed.(4)The effectiveness of the collaborative optimization model of speed guidance and signal timing at the reverse variable lane intersection is verified through simulation.taking a reverse variable lane intersection in Chongqing as an example,the simulation platform was established by using VISSIM COM interface,C# and MATLAB programming environment.Through designing three experimental schemes,the improvement effects under different permeability of connected automated vehicles and different traffic volumes were compared and analyzed.The results show that under the three traffic schemes,the improvement effects of the proposed co-optimization model of speed guidance and signal timing are better than that of the model with only signal optimization.It can significantly reduce the average intersection delay and fuel consumption of vehicles,and the improvement effects increase with the increase of the permeability of connected automated vehicles.Meanwhile,under the same permeability,the improvement effect of reverse variable lane intersection increases with the increase of left-turn vehicles.When the number of left-turn vehicles is large and the permeability of connected automated vehicles is high,the optimization effect of the intersection is more significant.