Research On Active Lane Changing Considering The Influence Of Cornering Stiffness

With the development of automatic control technology and computer technology,the level of intelligence of automobiles is getting higher and higher.Because of its vast prospects and the novelty it brings,it has always become a hotspot of research and attention.Among them,active lane changing is one of the key areas of smart car research.Research on the active lane changing of cars can help solve traffic congestion and improve traffic safety.However,the current research on active lane changing of smart cars mostly focuses on low-speed or good road conditions,and smart cars cannot avoid some highspeed or slippery road driving conditions as a means of transportation.Undoubtedly,the structural parameters of vehicles running at high speed on wet and slippery roads will change.Among them,the tire cornering stiffness,which is closely related to the tire lateral force,changes with road conditions,tire inflation pressure,vertical load,and other factors.If the tire cornering stiffness is set to a fixed value in the model-based control system,the uncertainty of the cornering stiffness will cause a larger error from the actual application.The performance of the trajectory tracking control depends on the accuracy of the control algorithm and vehicle parameters.Therefore,this paper conducts the following research on the active lane changing of smart cars considering the influence of cornering stiffness:First,the vehicle dynamics modeling is carried out,and Carsim is selected and the established model is used for simulation comparison and analysis,and it is verified that the established model can meet the design requirements of trajectory tracking control.Considering the influence of the surrounding vehicles,the safety distance is modeled to obtain the minimum safety distance;the fifth-order polynomial is used to plan the trajectory of changing lanes,and dynamic constraints are added to ensure the planned trajectory is smooth and safe.To solve the multiple constraint requirements of vehicles driving on low-adhesion roads,a trajectory tracking controller based on a multi-constraint model predictive control algorithm was designed,and its effectiveness was verified by simulation on low-adhesion roads.Then design the tire cornering stiffness estimation strategy,analyze the influence factors of the cornering stiffness,according to the selected low adhesion road conditions,ignore the secondary influence factors such as tire pressure and speed,and select the tire corner and vertical load as the main factor.This paper proposes a method of cornering stiffness estimation using a high-precision and high-efficiency extreme gradient boosting algorithm.At the same time,the radial basis function neural network algorithm is used to design the cornering stiffness estimation strategy as a comparison of the cornering stiffness estimation strategy based on the extreme gradient boosting algorithm.Since the lateral acceleration of the vehicle is close to the adhesion limit when the vehicle is driving at high speed on a slippery road,the tire lateral force can easily enter the non-linear region.The estimated cornering stiffness is added to the trajectory tracking control model.A trajectory tracking controller considering the change of lateral stiffness is designedFinally,Carsim and MATLAB/Simulink are used for joint simulation to verify the accuracy of the proposed tire cornering stiffness estimation and the performance of the trajectory tracking controller.To further verify the reliability of the research content,the relevant working condition tests were carried out on the hardware-in-the-loop platform of Carsim and Lab VIEW to verify the effectiveness of the trajectory tracking controller considering the influence of the cornering stiffness.