Research On Following Control Strategy Of Intelligent Vehicle Under Vehicle Infrastructure Cooperative Systems

With the rapid growth of global vehicle ownership,the problems of traffic accidents and traffic congestion have become increasingly prominent,which has become an important factor seriously restricting the healthy development of the national economy.Nowadays,the intelligent transportation industry is developing rapidly,the technology of vehicle coordination control has gradually become the main research hotspot of road traffic safety.It has attracted much attention as a solution to effectively eliminate the driver's instability and improve the efficiency of road traffic.Among them,intelligent vehicle as an important part of the man-car-road cooperative driving system,which is based on ordinary vehicles to add advanced sensors,controllers,actuators and other devices,through the vehicle-mounted sensor systems and information terminals to achieve information interaction with people,vehicles and roads,and has the real-time traffic environment perception,decision-making and control ability for vehicle safety,comfort,energy saving and other goals.Therefore,it is of great importance to realize the optimization of driving functions and establish a good road traffic environment value.This paper mainly focuses on the vehicle following process,and mainly considers the longitudinal control problems.The safety distance control model and the throttle / brake switching control model are established,and then the controllers are designed according to different control strategies.Meanwhile,the control effects are compared and analyzed through simulation.Finally,the experimental verification is completed based on intelligent vehicle system experimental platform under microenvironment.In this paper,the present situation of the development of vehicle road coordination technology at home and abroad is thoroughly understood,and the research situation of the following driving of the intelligent vehicle is summarized.Since Intelligent vehicle is the core of self-driving,the self-driving scheme is set up in this paper,the overall structure of the intelligent vehicle is designed according to this scheme,and the selections and functions of the vehicle sensors and the industrial personal computers are introduced in detail.The corresponding control systems are designed by the structure of underlying executive systems,which lays a foundation for the theoretical research and experimental testing of vehicle following control.Secondly,the expression of vehicle acceleration is established based on the vehicle dynamics model.According to the idea of fixed headway algorithm,the safety distance control model is proposed,and the control rules and control flow path of throttle / brake switching are designed.The overall control design of intelligent vehicle following system is completed.Then the research on the controller of following system is carried out,and the controller is designed by using fuzzy PID control and the BP neural network PID control strategy respectively.Finally,computer simulation is carried out by Matlab/Simulink software,the simulation results of relative distance error show that the accuracy of the fuzzy PID control is improved compared with the traditional PID control,and the BP neural network improves the control accuracy by using its self-learning ability,so that the effect of the BP neural network PID control is obviously better than that of the traditional PID control and the fuzzy PID control.Therefore,the algorithm of the neural network PID control is used to test.Due to the difficulty of the real scene test,the following experiments are carried out by using the test bench of intelligent vehicle system in miniature environment.The experimental results show that the BP neural network PID controller can reduce the relative distance error,and can meet the safety following driving of intelligent vehicles.