Robust adaptive decoupled-like sliding mode controller design based on iterative learning for overhead cranes

This paper proposes a novel model-free controller, considering the overhead crane systems cannot be accurately modeled and affected by external disturbances. The suggested scheme contains a variable exponential decoupled-like sliding mode control (VEDSMC), an adaptive power-reaching law (APRL), and a dynamic learning law-based iterative learning controller (DLLILC); they are combined in a parallel structure to form VEDSMC-APRL-DLLILC. The VEDSMC can effectively enhance the convergence speed of the displacement variables and improve the transient performance of the crane system. The APRL consists of an adaptive switching gain; it can estimate the optimal switching gain according to the unknown dynamics of the crane system and the disturbance, reduce the controller chattering, and guarantee the robustness. The DLLILC term can further improve anti-swing and positioning performance of the overhead crane without accurate information of the crane dynamics model in advance. Moreover, a nonlinear dynamic learning law (DLL) is developed to guarantee both convergence speed and steady-state accuracy in the learning process. Finally, the stability analysis of the designed controller is performed using Lyapunov theory and Barbalat's lemma, and the simulation results illustrate the effectiveness of the suggested control scheme.