| The rapid development of autonomous driving technology has enabled vehicles to operate autonomously without driver intervention.Nevertheless,the adhesion coefficient of road surfaces can decrease due to diverse weather phenomena such as rainfall and snowfall,leading to potential destabilization risks for vehicles utilizing conventional path tracking algorithms.In order to improve the path tracking accuracy and vehicle stability under the extreme conditions such as low adhesion,This thesis presents a novel model for joint path tracking and stability control that is based on a distributed independent drive electric vehicle platform.To ensure both precise path tracking and enhanced vehicle stability,lateral stability control is integrated into the path tracking control system.The research findings of this thesis can furnish novel insights and a theoretical foundation for enhancing the lateral stability of vehicles.Distributed independent drive electric vehicles possess the ability to exert precise and autonomous control over wheel torque,while simultaneously adjusting vehicular motion attitude and stability through the application of supplementary transverse swing moments.The present study addresses the path tracking problem for distributed drive vehicles by decoupling it into two distinct components: transverse path tracking control and longitudinal vehicle speed control.In this paper,the lateral path tracking problem is tackled through coordinate conversion based on the Frenet coordinate system.The error model is constructed utilizing the lateral and heading errors as state variables.To facilitate following a prescribed path,a prediction equation is derived using the MPC(Model Predictive Control)algorithm to determine the steering angle of the front wheels.Additionally,the mechanism responsible for generating tracking errors is analyzed,and a PID-based corner compensation strategy is formulated to mitigate these discrepancies.Furthermore,to address the longitudinal speed tracking problem,a longitudinal speed control model based on MPC is established.In this model,the target speed of the vehicle and its feedback speed deviation are employed as state variables.The longitudinal dynamics model of the vehicle is leveraged to solve for the desired acceleration speed,which in turn informs the calculation of the motor torque output via the driving balance equation.This approach results in a successful realization of longitudinal speed tracking.The present study addresses the lateral stability problem for vehicles operating on low adhesion roads.Initially,the destabilization mechanism of such vehicles is analyzed,and the lateral deflection angle of the center of mass and the angular velocity of transverse sway are introduced as critical indexes for evaluating their lateral stability.To overcome the challenge of directly acquiring the lateral deflection angle of the center of mass,an EKF(Extended Kalman Filter)observer based on the direct integration method is proposed to provide an estimation of this parameter.Subsequently,an STA(Super Twisting Algorithm)based transverse moment control model is developed based on the observed lateral eccentricity and vehicle state feedback parameters obtained from the observer.The performance of this model is compared against a SMC(Sliding Mode Control)model.Finally,a quadratic programming method is employed for distributing the resolved additional transverse moment to each wheel,thereby enhancing overall driving stability.Finally,the joint control model of path tracking and lateral stability is established through the joint simulation platform of CarSim/Simulink,and the effect of the model is verified under different road surface adhesion coefficients based on double-shifted road conditions,and the results show that the joint control model of this paper can effectively improve the path tracking accuracy and lateral stability of the vehicle under various low adhesion conditions. |
PDF Full Text Request