Research On Multi-Condition Optimization Of Adaptive Cruise Control System Based On MPC And ADRC

Nowadays,autonomous and driving assistance technology have attracted the attention of consumers and vehicle companies.As one of the important branches of this kind of technology,adaptive cruise system(ACC)has been widely studied and paid attention to.Due to the complexity of vehicle driving conditions and the uncertainty of vehicle parameters,the traditional ACC system controller with fixed parameters can not meet the comprehensive performance requirements under multiple conditions.Therefore,this paper studies the optimization of ACC system under multiple operating conditions.In this paper,a hierarchical control structure is used to design ACC system,and model predictive control(MPC)is used for the decision layer of ACC systems.The acceleration prediction model of the front car and the strategy of real-time adjusting the weight according to the different working conditions are proposed,which improves the model accuracy of the decision layer and the adaptability under multiple conditions.Then,a feedforward plus feedback controller based on the theory of vehicle dynamics and active disturbance rejection control(ADRC)is designed for the executive layer.Finally,the performance of the designed ACC control strategy is verified by simulation and real vehicle experiment.The research contents of this paper are as follows:Firstly,an overall control scheme of the ACC control system are designed,the whole system is divides into three parts,including the perception layer,the decision layer and the executive layer,and establishes the corresponding performance requirements.For solving the problem that the traditional constant time-headway distance(CTH)can not fully meet people's demand for the following distance under different vehicle speeds,several drivers' driving data are collected under the car-following conditions,and the collected data are extracted with characteristic parameters.In order to improve the car-following ability,safety and adaptability under unsteady car-following state,the forward vehicle speed term and acceleration term are introduced in the desired distance strategy,and then the stability analysis and verification of the designed modified distance strategy are carried out.Second,for considering the car-following ability,safety,fuel economy and passenger comfort of the ACC system comprehensively,the model predictive control(MPC)is used as the control algorithm.The decision layer output an optimal desired acceleration to act on the executive layer in each control period based on the current information and vehicle state.The constraint softening strategy based on relaxation factor is introduced to expand the feasible region of system solution.In order to improve the accuracy of the prediction model,the controller can still output the optimal control quantity when the acceleration of the front car fluctuates greatly,so the least square method is used to predict the acceleration at the future moment based on the previous acceleration data of the front car.Furthermore,the fixed weight strategy of cost function in traditional MPC can not meet the control demand because of the complex following vehicle condition and different driving environment,so a real-time adjustment strategy of weight is proposed according to the current distance error and speed error.Then,the executive layer of the ACC vehicle is designed.Firstly,the driving dynamics of the vehicle is analyzed,and the inverse longitudinal dynamics model is constructed as a feedforward control for vehicle driving and braking.As the parameters of the vehicle and the driving environment change,in order to improve the anti-interference of the lower controller and the accuracy of the output of the decision layer,the active disturbance rejection controller(ADRC),which does not depend on the precision model,is used as the feedback control of the executive layer,which enables the executive layer to track the desired acceleration of the decision layer stably and accurately.Finally,the control strategy of the designed ACC system is verified under multiple conditions and compared with the traditional MPC controller through Matlab/Simlink and Carsim simulation and real vehicle experiments,including steady-state following,front car cut-out,side car cut-in,uphill following and downhill following conditions.The safety,car-following,fuel economy and passenger comfort in the following process are analyzed.The results show that the proposed control strategy can have a good control effect under multiple conditions.