Composite anti-disturbance path tracking control for the unmanned vehicle under complex driving conditions

Abstract

The path tracking control of unmanned vehicles in complex driving conditions is frequently challenged by the model uncertainty and the external disturbance, such as the crosswind disturbance, which brings difficulties to the path tracking control. Therefore, to address the path tracking control of unmanned vehicles under complex driving conditions, a composite anti-disturbance path tracking control approach using multi-dimensional Taylor network (MTN) is proposed. First, based on the mechanism modeling of unmanned vehicles, the uncertainty is described by the MTN data-driven model; design the MTN disturbance observer to characterize the influence of the crosswind disturbance; the improved Back Propagation (BP) algorithm is proposed and treated as the learning algorithm of the MTN data-driven model and the MTN disturbance observer; the convergence of the MTN is proved. Then, design the MTN finite time controller, which tracks the reference path quickly and accurately based on principle of “feedforward compensation and feedback suppression”; according to the finite time control theory, the finite-time stability proof of closed-loop system is given. Finally, experimental simulations have been conducted, and the results show that the proposed approach can effectively track the path under complex driving conditions, and has good tracking performance, anti-disturbance performance, real-time performance, and strong robustness performance.