Road Traffic Flow Modeling And Control In Internet Of Vehicles

Vehicular networking technology establishes communication between driverless vehicles and communication between vehicles and infrastructure to form a huge network of information.Vehicular networking technology is not limited to apply to driverless vehicles,and it provides full-scale traffic information such as intelligent navigation,automobile working information and alert information to drive safety for manned vehicles.The emergence of vehicular networking technology provides a way to solve the modern traffic problems.It is difficult to solve complicated traffic problems,such as the serious phenomenon of traffic queuing at signalized intersection.So,it is imperative to apply the vehicular networking technology.Traditional traffic flow theory based on basic traffic parameters is not suitable for vehicular networking technology.Furthermore,the traffic flow model of driverless vehicles is limited.The intelligent controllability of the driverless vehicles makes it possible to model to discover complex nonlinear phenomena of the traffic and apply classic feedback control theory in internet of vehicles to make the traffic flow orderly in road section.Those play an active role for the realization of warning and the realization of the orderly traffic in internet of vehicles.The contents of the paper are as follows:In open boundary conditions,the establishment of modified coupled map model without overtaking in single-lane based on full velocity difference model is raised up.Furthermore,in order to effectively adjust traffic flow in road section to ease traffic,a decentralized delayed-feedback control method for driverless vehicles is put forward in internet of vehicles.And applying the control method to coordinate signal control at signalized intersection,the effect of green wave band will come true better.The modified discrete dynamic coupled map model is derived from both the flow-density-speed fundamental diagram and Del Castillo's speed-density model.The modified model employs occupancy as its new variable and introduces a coupling strength with the consideration of effect of the front adjacent vehicle.And a procedure to design the decentralized delayed-feedback controllers is provided for the traffic control system in internet of vehicles.For the realization of warning of lane change in internet of vehicles,a interrupted traffic flow model is calibrated and validated with a nonlinear optimization procedure LevenbergMarquardt Algorithm using the trajectory datasets provided by Next Generation SIMulation(NGSIM)program to accurately describe the real vehicle lane changing in reality.For the realization of warning other vehicles its the direction of travel at signalized intersection in internet of vehicles,the characteristics of traffic flow which include the process of parking and dissipation are studied.The changes in velocity and time-headway of traffic flow are analyzed.And a control method for driverless vehicles in internet of vehicles is proposed.The control method of alleviating expressway congestion situation in internet of vehicles is studied.In automated highway system,two different methods of feedback linearization for single road section and for whole road section are applied.Numerical simulations show that the control method can keep the density and speed to the desired state and make highway system stable and orderly.