Collaborative Control Strategy For Mixed Traffic Flow Near A Highway On-ramp

Traffic jam phenomenon at the on-ramp bottleneck on highways has become a new normal,which reduces traffic capacity and brings inconveniences to people’s daily travel.Ramp metering and variable speed limit control,as the most commonly used traffic management methods,adjust on-ramp and mainline flow entering the bottleneck area,to as much as possible reduce traffic congestion and improve traffic efficiency of highway system.However,at the area around an on-ramp,many factors should limit existing traffic flow control measures,such as the inherent merging conflicts between mainline traffic and on-ramp traffic,the sudden increase in the number of unpredicted lane changes,various driving characteristics of different vehicles and the limited accuracy of fixed detectors.In recent years,the development of advanced communication technology and sensing technology has promoted more intelligent and connected vehicles.This indicates that mixed traffic flow composing of manual driving vehicles and connected and automated vehicles(CAVs)will become the trend of transportation system in the future.Also,it is an opportunity to improve existing traffic flow control methods.This paper takes on-ramp bottleneck of a highway as object of study,focusing on state evolution process of mixed traffic flow at the on-ramp bottleneck.A collaborative control strategy is proposed with optimal traffic flow control and state equilibrium of local area.The traffic simulation method is used to simulate operation of mixed traffic flow.Traffic flow characteristics analyzed,an optimal control strategy is developed.Besides,a strategy is presented for state equilibrium between lanes at a local area,to make traffic flow pass the merging area stable and orderly.The control method can improve traffic efficiency and safety at the same time,to ensure higher service efficiency of highway system.Firstly,in this paper,the dynamic evolution of mixed traffic flow at the on-ramp bottleneck is analyzed.The merging conflict between mainline flow and ramp flow is one of the main reasons for traffic congestion here.Merging maneuver of ramp vehicles should increase deceleration operations of upstream mainline vehicles,which may give rise to traffic disturbances.What is worse,too small merging gaps tend to raise emergency braking of mainline vehicles,thereby causing collisions or traffic jams.In order to develop a mixed traffic simulation environment,car following models and lane changing rules are utilized to simulate driving behavior of manually driven vehicles and CAVs.In this environment,cooperative merging of CAVs can promote smooth interactions between mainline traffic and on-ramp traffic.Nevertheless,merging operations of ramp flow may still interfere with upstream traffic on the main road,especially when penetration rate of CAVs is low.The traffic breakdown happens near a merging area will extend longitudinally and laterally,with traffic capacity decreasing.Existing traffic flow control methods are scarcely able to deal with heterogeneous traffic flow and traffic fluctuations.A framework of hierarchical control is put forward in this paper for different areas of highways.The optimal control of mixed traffic flow combines variable speed limit control and ramp metering,considering traffic characteristics and variations of highway sections with an on-ramp.Based on traffic data collected by fixed detectors and real-time state information sent by CAVs,the METANET model and the unscented kalman filter algorithm are integrated to estimate current traffic condition and predict traffic state in a certain time horizon.The minimum of travel time and speed variation is taken as the objective of optimization control of traffic flow.With genetic algorithm solving this problem,the value of variable speed limit on main road and the ramp metering rate are determined together.Compared to ALINEA ramp metering and a coordinated strategy of variable speed limit control and ramp metering that optimizes traffic efficiency,the proposed method predicts traffic disturbances and variations and reduces the adverse effect of merging traffic on speed of mainline traffic.Correspondingly,traffic efficiency is increased and possibility of traffic breakdown is dropped.From a macro perspective,smooth operation of traffic flow in bottleneck areas is the key to stable operation of highway corridors and even the entire transportation system.The state equilibrium strategy is developed for traffic flow of several lanes in a local area.It includes the lane changing guidance strategy at the upper boundary of the control area and the cooperative merging control strategy for vehicle state adjustment.The former is for mainline vehicles at the rightmost lane and the latter considers both main road and on-ramp vehicles.In the merging section,a roadside control unit is specially set up.Through advanced sensing technology,the unit is able to obtain real-time information of traffic situation.For example,current position and speed of vehicles in the control area can be collected by a Li DAR.At the same time,via wireless communication technology,state information of CAVs is received continuously.The lane changing guidance signal sends control command to vehicles at the specified position,to adjust number of vehicles entering different lanes downstream.Meanwhile,the collaborative merging strategy takes into consideration driving characteristics and constraints of mainline and on-ramp vehicles,and takes stability of platoon passing the merging point as the optimal objective.The genetic algorithm is used to determine the optimal vehicle merging sequence,with desired vehicle trajectories determined correspondingly.As many uncertain factors in real road may have influence on safety and stability of platoons,the model predictive control method is adopted during strategy implementation to adjust control schemes in real time.In the case of relatively high penetration rate of CAVs,the integration of lane changing guidance and cooperative merging control should ensure that traffic flow through the merging area runs smoothly and efficiently in a desired state.To verify effect of different levels of control strategies on mixed traffic in various scenarios,a simulation test system is built based on basic environment of mixed traffic flow.The modules of control system and simulation evaluation are added to analyze traffic flow performance indicators under different levels of control methods.The capabilities of the hierarchical control can be judged in improving traffic operation efficiency and alleviating traffic congestion.At the same time,adaptability of the proposed method to different mixed traffic flow environments is verified.