Nonlinear vibration suppression control of underactuated shipboard rotary cranes with spherical pendulum and persistent ship roll disturbances

Abstract

Shipboard rotary cranes are widespread used in industrial fields to carry out large-scale cargoes under marine environment, which are a class of representative nonlinear systems. For crane systems, there exist unactuated state variables (e.g., payload swing angles), which hence make them typically underactuated systems, whose degrees of freedom (DOFs) are more than the number of control inputs. Besides, due to the complexity disturbances of the marine environment, there are always inaccuracies in the three-dimensional (3D) dynamics model of shipboard rotary cranes, such as ocean waves, wind waves, ocean currents and so on, which make the control methods hard to achieve the perfect control effect. Most of the existing methods are built on the basis of simplified linear model, which often require precise model parameters or velocity signals. In order to solve the control problems caused by the above uncertainties, this paper proposes two nonlinear anti-swing control methods. First, considering the uncertain parameters, an adaptive controller is designed, which can achieve the control effect of suppressing swing when the model has unknown parameters. Second, considering the unmeasurable velocity signals, an output feedback controller is designed, which can achieve the control effect to suppress the swing angle with no need of velocity signals. On the basis of the above two control methods, we design corresponding different energy storage functions including kinetic energy and potential energy, respectively. Specifically, this paper has achieved an excellent effect of restraining swing without any linear approximation of the dynamic model. Lyapunov theorem and LaSalle invariance principle are utilized to further prove the asymptotic stability of the closed-loop system. Finally, experiments are implemented to further demonstrate the effectiveness of the proposed control methods.