ABSTRACT To address the challenges in finite‐time filtered tracking control of uncertain unmanned aerial vehicle systems with position‐attitude coupling under external disturbances and input saturation, an adaptive finite‐time anti‐saturation controller with disturbance compensation is proposed. First, to address the inherent complexity of repeated differentiation in the backstepping method, a filtering‐based control technique is adopted, which not only reduces differential computations but also simplifies the controller structure. Second, the universal approximation capability of neural networks is exploited to handle uncertainties arising from aerodynamic drag‐induced parameter variations, modelling errors, and other unknown dynamics. Third, a disturbance observer is designed and incorporated into the controller to compensate for the effects of external disturbances on the system. Moreover, a hyperbolic tangent function is employed to address the non‐differentiability issues inherent in conventional saturation functions. Furthermore, a finite‐time control approach is adopted to guarantee the UAV system's stability within a finite time. Finally, the effectiveness, robustness, and superiority of the proposed control scheme are demonstrated through numerical simulations.
Yang et al. (Thu,) studied this question.