Adaptive Robust Path Tracking Control for Autonomous Vehicles Considering Multi-Dimensional System Uncertainty

As the bottom layer of autonomous vehicle, path tracking control is a crucial element that provides accurate command to X-by-wire chassis and guarantees vehicle safety. To overcome deterioration performance for path-tracking controllers caused by modeling errors parameter perturbation, an adaptive robust framework proposed in this paper. Firstly, 2-DOF dynamic model established non-singular fast terminal sliding mode algorithm adopted formulate law. The unmeasured disturbance perturbation regarded as system uncertainty. enhance accuracy, radial basis forward neural network introduced estimate such uncertainty real time. Then, active front steering established. reference applied torque considering brought dissipation manufacturing mechanical wear. Finally, Simulink–CarSim co-simulation platform used validated two test scenarios. simulation results demonstrate has satisfactory good robustness against