Research On Vehicular Adaptive Cruise Control On Curved Road

In recent years,traffic accidents caused by human factors are increasing.The further development of active safety technology is imminent.Adaptive Cruise Control(ACC)is one of them.In order to improve the level of vehicle automation,alleviate the driver’s driving burden,reduce driver’s misoperation as well as ensure the driving safety,operation stability and traffic efficiency,this thesis focuses on vehicular ACC system on a curved road.First,the overall control scheme is designed at different levels and effective target identification algorithms,control strategies and execution structure are designed respectively.Based on the perceptual layer data and geometric relationships,turning and lane change identification is carried out for the front vehicle.What’s more,the identification model for adjacent and self-lane vehicles on structured curve road is designed to distinguish effective tracking targets.The safety distance models are analyzed in this paper and the amendment design for the relative distance error of radar detection after the front vehicle enters the curve is carried out.Next,the lateral and longitudinal control methods are respectively carried out for the curved road condition.Active Disturbance Rejection Control(ADRC)is applied to the lateral control of the vehicle to achieve the steering following efficiency.For longitudinal control,different control strategies are adopted according to the different traffic environment.When the vehicle is in a free flow state,the free flow speed control is designed combined with the road conditions.At this time,the Proportional-Integral-Differential(PID)controller is used to control the vehicle speed;when the vehicle is in the non-free flow state,the non-linear fuzzy control rules are formulated to realize the smooth car-following.Then,the executive layer of the ACC system is designed.Combined with the knowledge of vehicle system dynamics,a vehicle driving and braking control module based on an inverse dynamic model is constructed.Due to the strong non-linearity of the vehicle and the changeable environment during driving,the exact model is too complex to calculate and the simplified model is not matched sufficiently.Taking "feedforward + feedback" control to enhance the execution stability.The feedforward control is designed according to the driving dynamics and inverse dynamics models,and the ADRC with low dependence on the model accuracy and strong anti-interference is introduced as feedback compensation.Then the implementation effect of the controller is verified.Finally,the curved ACC system designed in this paper is simulated based on Carsim simulation platform and MATLAB/Simulink,including front vehicle turning identification,the recognition of the position relationship of the vehicle in front,relative distance compensation effect in the corner,self-driving lateral control as well as longitudinal control.The simulation results show that the ACC system designed in this paper can accurately identify the effective target vehicular state and achieve favorable control effect under different working conditions.