Research on Robust Model Predictive Control Based on Neural Network Optimization for Trajectory Tracking of High-Speed Autonomous Vehicles

Abstract

A robust model predictive control method for trajectory tracking is proposed to achieve higher trajectory tracking accuracy and stability of autonomous vehicles traveling at high speed. To reduce the interference of the external road environment, vehicle longitudinal speed, and the nonlinear characteristics of tire cornering stiffness on the vehicle model, a convex polytope vehicle dynamics model considering tracking error is adopted to improve the robustness of the vehicle model. The robust trajectory tracking controller is designed by combining the trajectory tracking multi-objective constraints and polytope model, while the correction term feedback is introduced to improve the robustness and control accuracy of the control system. A projection neural network solution model is established to optimize the objective function and reduce the occupation of computing resources. Simulation results show that compared with the traditional model predictive control, the control method designed in this paper has better trajectory tracking and that the vehicle is more stable at high speed.