Vehicle Adaptive Cruise Control Based On Rolling Optimization

Adaptive cruise control(ACC)system has received wide attention of scholars and automobile manufacturers in recent years,Because it can replace the driver for longitudinal control of the vehicle and can also reduce traffic accidents caused by human factors.Aiming at the current ACC system of comfort and fuel economy is not adequately considered in the design process,And can only complete the specific conditions of cruise or car following,This paper will consider safety,traceability,comfortable and fuel economy under a Model Predictive Control(MPC)framework in the process of vehicle controller design.in order to increase the adaptability of road traffic,The method of variable weight coefficient of the different conditions is put forward.At the same time multi-mode ACC system will also be studied.Due to the direct structure may affect the robustness of the system,Therefore,this paper uses the hierarchical control scheme to complete the design of car-following system.First,the expected two vehicle spacing is calculated according to the strategy module,then upper controller decisions out the desired longitudinal acceleration,according to the vehicle information collected by the radar sensor,The lower controller is converted into the corresponding accelerator pedal or the brake pedal according to the inverse model of the vehicle.The actuator unit will transmitted the command to the vehicle,Then the control flow of the whole system is completed.For the upper controller need to be considered multi-objectives,including safety,traceability,comfortable and fuel economy,the actual constraints should also be considered,Therefore,this paper adopts model predictive control(MPC)method which can be used to deal with multi-objectives and system constraints;For the lower controller requirements the actual acceleration tracks the expectations fastly,therefore,using feedforward and feedback control method.Feedforward control is helpful to eliminate the influence of the nonlinear term in the system,and the feedback control adopts PID control algorithm,which can meet the requirements of the system.This paper puts forward a method of variable weight coefficient,In order to increase the ACC system’s ability to adapt to the environment.Fuzzy theory is used to identify the weight coefficient online of car-following and comfort in the objective function according to the different driving conditions,to achieve focuses on different purposes under different working conditions.In order to verify the validity of the designed hierarchical control strategy,According to five typical traffic scene off-line simulation validation respectively.The results have shown that tracking performance is good in different scenarios,the vehicle can take into account multi-objectives in the process of driving at the same time.The vehicle states also have a smooth transient response.The information of the road environment is complex and changeable,and various factors may affect the driver’s decision,Due to the parameters of one controller are fixed,thus limiting the use of ACC system.For this,this paper also researches on the multi-mode ACC system.According to the emergency degree coefficient,The car-following conditions are divided into cruise,accelerate approaching,steady following,decelerate approaching and emergency following,Corresponding to each condition,adjust the weight coefficient of the objective function and the corresponding constraints complete each mode controller.In order to verify the effectiveness of multi-mode ACC system,five typical driving conditions are selected firstly,then set two extreme driving conditions,The simulation results show that the multi-mode ACC system has good tracking performance in typical scenarios,For the extreme conditions,Multi-mode ACC system can take corresponding braking measures as early as possible to avoid collision accidents,At the same time it can take into account the multi-objectives of the vehicle,and has certain controlling smoothness.In this paper,the research is based on the vehicle ACC system,under the framework of MPC the upper level controller is designed,which is based on the higher order model.The algorithm used in the lower level controller is feedforward and feedback for vehicle inverse longitudinal dynamic model,At the same time,in order to increase the adaptability of road traffic,this paper puts forward the method of variable weight coefficient.The idea of multi-mode ACC system is put forward In order to enlarge the application scope of ACC system,The simulation results show that the two schemes have good tracking performance and dynamic response in typical scenes,Under the limit condition,multi-mode ACC system can avoid collision accident through emergency braking,thus expanding the scope of application of ACC system.