Longitudinal Control Of Vehicular Platoon Under Mixed Traffic Environment

With the increasing number of motor vehicles coupled with the slower expansion of the road network,traffic congestion and traffic accidents have become a serious problem in today's society.In recent years,the rapid development of CAV(Connected and Automated Vehicle)technology has provided a feasible way to solve this problem.Platooning of CAVs in roads enables each vehicle to drive in a smaller spacing,which can greatly improve the utilization rate of roads and effectively avoid collisions.This is an inevitable trend in the development of future intelligent transportation.However,with the continuous development of CAV technology,vehicular platoons are likely to be composed of a mix of HDVs(Human-driven Vehicles)and CAVs in the near future.This complex traffic environment will bring huge challenges to traffic flow modeling,control,and management when considering the stochastic driving characteristics of humans and the uncertainty of the interaction between CAVs and HDVs.Therefore,how to build a fleet control model under the coexistence of CAVs and HDVs and improve the safety of the mixed vehicular platoon will be an important research topic that needs to be urgently addressed.To this end,this paper takes the mixed vehicular platoon as the research object and adopts the hierarchical control architecture to design the longitudinal control system of mixed platoons.Firstly,the vehicle dynamics are reasonably simplified according to the vehicle's driving conditions.Secondly,the different characteristics of car-following behaviors of CAVs and HDVs are analyzed and compared,and the longitudinal mathematical model of the mixed platoon is established based on the virtual MSD(MassSpring-Damper)theory.Then,the theory of virtual energy field of mixed platoons is proposed.This theory is used to describe the spacing errors and speed errors of the vehicles during the car-following process,and the magnitude of the virtual total energy serves as a reference index to objectively reflect the danger of the car-following process.Finally,a driver-in-the-loop test platform is established and the effectiveness of the proposed mixed platoon control algorithm is verified.The main research contents of this paper are as follows:(1)Vehicle dynamics modeling and reasonable simplification based on the vehicle's driving conditions.First of all,the vehicle's movement status and force are analyzed according to the actual driving conditions of the mixed fleet,and a vehicle dynamic model consisting of an engine,a torque converter,a drive train,tires,and brakes is established.In addition,because the vehicle has a high degree of non-linearity,it is necessary to simplify the model to ensure the accuracy of the vehicle model while reducing the difficulty of designing the control system and improving the real-time nature of the control system.Finally,the reasonably simplified dynamic model is simulated under CarSim/Simulink environment to verify the accuracy of the model.(2)Establish a longitudinal mathematical model of the mixed vehicular platoon based on the virtual MSD system.First,according to the coexistence of HDV and CAV in future road traffic,the different characteristics of the car-following behaviors of CAVs and HDVs are analyzed and compared.Then,considering the spacing errors and speed errors of the mixed fleet during the driving process,a mixed vehicular platoon model is established based on the virtual MSD theory.Finally,according to the established MSD mixed platoon model,the mathematical conditions to ensure the stable driving of the fleet are proposed,so that the spacing errors of the vehicle during the driving process is not amplified when transmitting to the rear of the fleet.(3)Construct the minimum energy performance function of the mixed platoon based on the psychological field theory.Driving behavior is the result of a driver's judgment based on their psychological expectations when being stimulated by external environmental information.Every driver will have a psychological field during the driving process.The changes in the external traffic environment will exert forces on their psychological field,thereby adjusting the speed and direction of the vehicle.Therefore,based on the psychological field theory and the MSD model of the mixed vehicular platoon,a virtual energy field of the mixed vehicular platoon is firstly established at the system level.Then,a safety evaluation index is designed based on the established virtual energy field to objectively reflect the danger of the car-following process,and the abnormal driving behavior in the mixed traffic flow can be accurately identified and predicted based on this index.Finally,the driver's psychological change during the carfollowing processes is expressed as the energy accumulation of the MSD system,and the minimum total energy control method for a mixed vehicular platoon is proposed.At the same time,a controller of the mixed vehicular platoon is designed based on vehicle performance and dynamic constraints.(4)Build a driver-in-the-loop test platform and perform simulation tests under typical driving conditions.The Logitech G29 vehicle controller,PreScan,and Matlab/Simulink are used to build a driver-in-the-loop test platform.On this platform,a typical driving scenario of a mixed fleet is set up,and drivers are randomly selected to simulate HDVs in the fleet to verify the effectiveness of the mixed platoon control algorithm.In addition,the dominant factors influencing the driver's car-following behavior are analyzed based on the minimum total energy in the platoon.