A Robust Dual-Loop Control Framework for Quadrotor UAV Trajectory Tracking

Abstract

Quadrotor unmanned aerial vehicles (UAVs) are characterized by highly coupled, nonlinear, and underactuated dynamics, which make accurate trajectory tracking challenging in the presence of external disturbances, actuator faults, and model uncertainties. To address this problem, this paper designs a robust dual-loop trajectory tracking control framework. In the position loop, a fractional-order nonsingular fast terminal sliding mode controller (FONFTSMC) is designed to improve transient convergence and steady-state accuracy. In the attitude loop, an adaptive radial basis function (RBF) neural network is incorporated into the fractional-order sliding mode controller to approximate unknown model uncertainties, while adaptive gain estimation and robust compensation terms are introduced to handle actuator effectiveness loss, bias faults, approximation residuals, and external disturbances. Three-dimensional spiral trajectory simulations demonstrate that the proposed method achieves accurate tracking, bounded attitude responses, and improved robustness compared with NFTSMC.