Data-driven discrete learning sliding mode control for overhead cranes suffering from disturbances

Currently, due to the increasing complexity of industrial processes, obtaining precise dynamic models for cranes is challenging. Cranes often operate in complex environments and are inevitably affected by uncertain disturbances, which can jeopardize control performance and stability. Therefore, this paper introduces a data-driven approach that frees itself from reliance on the system model by estimating the system model T using input–output data. The proposed discrete sliding mode learning control utilizes a learning term in place of the traditional discrete sliding mode switching term, thereby eliminating inherent chattering in traditional discrete sliding mode and enhancing the system’s convergence rate. In cases of uncertain disturbances, a disturbance observer based on the output is employed to estimate and compensate for these disturbances, ensuring robustness in system control performance. The effectiveness of the proposed controller is validated through multiple comparative experiments.