Adaptive neural network-based super-twisting sliding mode control for UAV trajectory tracking under disturbances

This study investigates the fast-tracking control of a quadrotor system operating in a perturbed environment with uncertain states, employing a novel control strategy. The proposed method integrates adaptive super-twisting nonsingular terminal sliding mode control with a multi-layer neural network to effectively address the quadrotor's tracking control. The control strategy is implemented in two phases. In the first phase, the quadrotor's position is assumed to be unknown, and a multi-layer neural network is employed to determine the position. The second phase focuses on the tracking control of the quadrotor system under external disturbance, utilizing an adaptive super-twisting nonsingular terminal sliding mode control to ensure fast-tracking control while rejecting external disturbances. In both phases, the stability of the closed-loop system is established using the Lyapunov theory. The proposed method ensures accurate estimation of the quadrotor's position through the multi-layer neural network, promotes rapid convergence of the proposed super-twisting nonsingular terminal sliding mode surface, provides superior and significant chatter-free control, and introduces an adaptive mechanism to estimate the uncertain bounds of external disturbances. To demonstrate the effectiveness and performance of the control technique, extensive evaluations are conducted via simulations using MATLAB/Simulink, comparative analysis with existing techniques, and hardware-in-the-loop implementation with Speedgoat hardware.