Trajectory Planning and Control of Autonomous Vehicle Under Extreme Conditions Based on Autonomous Drift

To consider both stability and trajectory tracking performance of autonomous vehicles operating in extreme conditions,a trajectory planning and control method based on autonomous drift is proposed. A neural network tire dynamics model is designed based on neural network to improve the accuracy of the traditional magic tire formulation. In order to further expand the stability boundaries under the extreme working conditions of autonomous vehicles,the drift stability boundaries are designed based on the tire saturation and maximum sideslip characteristics combined with the center-of-mass lateral deflection angle-transverse swing angular velocity phase plane constraints during drift,and the nonlinear model predictive control(NMPC)is used to plan a safe drift trajectory within a wider stability range,and the drift tracking control is carried out for the planned trajectory. The results of the joint simulation of Simulink/CarSim show that the method can fully utilize the advantages of drift motion to ensure that the vehicle does not go out of control under extreme working conditions,while accurately tracking the desired trajectory. © 2024 SAE-China. All rights reserved.