Research On Longitudinal Platoon Control Considering Information Flow Topology

Vehicle platoon control is a hot research topic of intelligent transportation system.On the condition of ensuring safety and stability,shortening the distance between vehicles that can improve road traffic capacity,effectively alleviate road traffic congestion and reduce vehicle fuel consumption.Communication,as the key of realtime information interaction in vehicle queue,has a very important impact on the performance of vehicle control system.In this paper,the influence of topological structure of information flow on the control performance of homogeneous vehicle queue was discussed.A vehicle platoon model based on nonlinear longitudinal dynamics of vehicle is constructed by using the selected topological structure of information flow,based on terminal constraints and Lyapunov stability theory,the corresponding model predictive controller is designed and simulated by Matrix Laboratory(MATLAB)platform,the validity and stability of the controller under the selected topology of information flow are verified.The main researches of this paper are as bellowing:(1)Firstly,as the foundation of platoon models in Chapter 3 and Chapter 4,the vehicle kinematics model and the nonlinear vehicle dynamics model which based on vehicle dynamics theory are established.Secondly,the author also described the existing spacing strategy,control objective and topology structure of information flow,and introduced the model predictive control theory and stability theory,which lays a foundation for the design of the controller.(2)Firstly,the linear homogeneous vehicle platoon model is established which is based on kinematics model,and a distributed model predictive controller is designed under each information flow topology structure.Secondly,this paper describes the optimal control problem of vehicle platoon control system in finite time domain,and transforms it into linear quadratic problem to solve it.Finally,the simulation is carried out under Matrix Laboratory(MATLAB)platform,and the results show that the convergence effect of the controller is better under each information flow topology structure.(3)The heterogeneous vehicle platoon model based on vehicle nonlinear dynamics model is established,and the corresponding predictive controller of heterogeneous vehicle platoon model is designed under the information flow topology structure of the lead vehicle.By using terminal constraints and Lyapunov stability theory,the sufficient conditions for the controller to meet the asymptotic stability are obtained,and the weight matrix is adjusted to ensure the asymptotic stability of the vehicle platoon system.(4)According to the typical working condition in Matrix Laboratory(MATLAB),the vehicle queue model predictive control system under the information flow topology structure of the leading vehicle is simulated and analyzed.The simulation results show that: under the information flow topological structure simulation of the lead vehicle,the vehicle fleet control shows better robustness;The calculation time of pilot-vehicle following is half of that of pilot-double-vehicle following,so the former is more suitable for real-time control in practical application,the algorithm proposed in this paper has a maximum single-step solution time of less than 0.2s sampling period,the real-time requirement is satisfied in every working condition.