Sliding-mode active disturbance rejection control for container cranes based on multi-source disturbance adaptive bandwidth disturbance observer

Abstract

To address the performance degradation in positioning and anti-sway control of four-rope container cranes induced by multi-source port disturbances, a composite sliding mode active disturbance rejection control scheme integrated with a Multi-Source Disturbance Adaptive Bandwidth Disturbance Observer (MSD-AB-DOB) is proposed. First, external disturbances are categorized into three distinct classes—trolley actuation, load swing coupling, and inter-rope coupling—to resolve the insufficient discrimination of disturbance characteristics prevalent in existing studies; subsequently, independent dynamic models are formulated, and three parallel sub-observers are developed for hierarchical decoupled estimation of these multi-source disturbances. Second, an adaptive bandwidth mechanism driven by measurable signals is introduced to mitigate the inherent trade-off between measurement noise sensitivity and estimation lag associated with conventional fixed-bandwidth observers. Concurrently, an Extended State Observer (ESO) is employed to independently estimate internal lumped disturbances, such as model uncertainties. Furthermore, sliding mode control (SMC) is utilized to reject residual compensation errors. Consequently, a hierarchical composite architecture characterized by “external disturbance observation–feedforward compensation–internal disturbance suppression–sliding mode robustness” is established to sequentially attenuate both internal and external disturbances. The uniform ultimate boundedness of the closed-loop system is rigorously proven via Lyapunov theory. Simulation results demonstrate that the proposed method outperforms traditional fixed-bandwidth DOB and standalone SM-ADRC schemes, enabling precise positioning and effective sway suppression while significantly enhancing overall control performance.