Adaptive Disturbance Stability Control for Uncrewed Aerial Vehicles Based on Radial Basis Function Neural Networks and Backstepping Sliding-Mode Control

This paper focuses on the stability control of unmanned aerial vehicles (UAVs) under external disturbances. UAVs are frequently subjected to external disturbances, with a significant impact stemming from wind effects, which adversely affect the stability of trajectory tracking control. Current research on UAV control has not effectively addressed these external disturbance issues; while some controller algorithms may exhibit good suppression capability under minor disturbances, their effectiveness decreased significantly as the magnitude of disturbances increases. In order to enhance the interference immunity of the controller, this study proposes a method based on Radial Basis Function Neural Networks (RBFNN). The RBFNN is employed to estimate external disturbances, and an adaptive weight adjustment algorithm is utilized to accommodate various types of interference. This approach effectively suppresses disturbances by integrating the backstepping sliding mode control method into the design of both position and attitude controllers. This combined method not only enables real-time estimation of complex disturbance signals but also facilitates rapid adjustments to the control strategy, ensuring stable flight control. Simulation experiments demonstrate that the proposed control scheme maintains excellent trajectory tracking and attitude control performance under various disturbance conditions. Under specified disturbance conditions, the proposed controller exhibits strong anti-disturbance capabilities in both trajectory and attitude angle tracking, maintaining a tracking error of less than 1% under all applied disturbances, and outperforms other advanced controllers in overall performance.