Neural Network‐Driven Adaptive Control for Swing‐Free Trajectory Tracking in Double‐Link Overhead Cranes With Uncertain Dynamics

ABSTRACT Overhead cranes are widely used in industry owing to their capability of handling heavy payloads and providing flexible operation. However, they are inherently underactuated systems with complex dynamics, subject to system uncertainties and external disturbances during operation. Consequently, achieving high‐performance control while ensuring safe payload transportation remains a challenging task. To address these challenges, this paper proposes a novel control framework that integrates radial basis function neural networks (RBFNNs) with adaptive sliding mode control (ASMC) for three‐dimensional overhead crane (3DOC) systems with double‐link dynamics, aiming to enhance trajectory tracking accuracy and suppress payload oscillations during motion. Compared with existing methods, the proposed approach overcomes fundamental limitations in overhead crane control by employing RBFNNs for accurate estimation of system uncertainties while exploiting the robustness of ASMC against external disturbances and approximation errors. The stability of the closed‐loop system is rigorously established using Lyapunov theory. Finally, simulation studies on a quasi‐physical model, with comparative evaluations against state‐of‐the‐art controllers, demonstrate the feasibility and superior performance of the proposed controller in achieving precise trajectory tracking and effective payload oscillation suppression.