Research On Multi-Objective Control Of Adaptive Cruise Control System Under Curve Condition

The adaptive cruise control system(ACC)can reduce the driver’s workload and improve the vehicle’s driving safety by taking over the longitudinal control of the vehicle.At present,with the development of vehicle sensors and vehicle networking technology,the ACC system can be extended to be used under the curve conditions.Different from the traditional ACC system,steering operation,higher driving speed,acceleration/deceleration and other factors will increase the risk of lateral instability of ACC vehicles in the process of high-speed car following on curves.Although the direct yaw moment control(DYC)system can effectively improve the lateral stability of vehicles by applying different longitudinal forces on different wheels,ACC may interact with it under some conditions during the intervention of DYC system.In this thesis,the dynamic model of ACC vehicle and the kinematics model of carfollowing are established.Then,the prediction model under MPC framework is established and deduced by using the above ACC vehicle dynamics model and kinematics model.This thesis analyzes the conflicts and contradictions between various performance indexes of ACC vehicle when it follows the vehicle in the curve.For example,under some conditions,the acceleration/deceleration limit of ACC vehicle will make the carfollowing performance of ACC worse.When the acceleration of ACC vehicle is limited,the car-following distance between the host vehicle and the preceding vehicle may become larger,resulting in frequent cut-in of the rear vehicle;when the deceleration of ACC vehicle is limited,the risk of rear end collision may occur.Therefore,the carfollowing performance and longitudinal ride comfort of ACC vehicles are contradictory and restricted under some conditions.In order to solve the contradiction and conflict between ACC vehicle performance in the process of car-following,the extension control method which can coordinate and solve the contradiction problem is introduced into the weight design of each performance index under MPC framework,so as to coordinate each performance of ACC vehicle.In order to solve the problem that the traditional constant-weight MPC cannot adapt to various complex conditions,this paper uses extension control to divide the extension area according to the state of ACC vehicles.Under different vehicle states and longitudinal car-following states,the corresponding real-time weight matrix is designed by using the correlation function of extension control method,so as to further coordinate various control objectives and improve the overall performance of curve ACC performance.The performance of ACC vehicle with different weights is simulated and analyzed by using the co-simulation platform of Car Sim and MATLAB / Simulink.The simulation results of the traditional ACC method,the traditional ACC&DYC method and the extension variable weight control method proposed in this paper are compared and analyzed to verify the effectiveness of the proposed method.Finally,in order to verify the effectiveness of the proposed extension variable weight curving ACC,the driver-in-the-loop(DIL)test is carried out by driving simulator.The test results show that the proposed method can effectively improve the overall performance of ACC vehicles,achieve the coordination of various control objectives,and improve the longitudinal ride comfort and fuel economy under the premise of ensuring the car-following performance and lateral stability.