Neural Network-based Robust Anti-sway Control of an Industrial Crane Subjected to Hoisting Dynamics and Uncertain Hydrodynamic Forces

In this paper, a neural network-based robust anti-sway control is proposed for a crane system transporting an underwater object. A dynamic model of the crane system is developed by incorporating hoisting dynamics, hydrodynamic forces, and external disturbances. Considering the various uncertain factors that interfere with accurate payload positioning in water, neural networks are designed to compensate for unknown parameters and unmodeled dynamics in the formulated problem. The neural network-based estimators are embedded in the anti-sway control algorithm, which improves the control performance against uncertainties. A sliding mode control with an exponential reaching law is developed to suppress the sway motions during underwater transportation. The asymptotic stability of the sliding manifold is proved via Lyapunov analysis. The embedded estimator prevents the conservative gain selection of the sliding mode control, thus reducing the chattering phenomena. Simulation results are provided to verify the effectiveness and robustness of the proposed control method.

Automated summary

A neural network-based robust anti-sway control method is proposed for a crane system transporting an underwater object. By embedding estimators in the system to compensate for uncertainties and unmodeled dynamics, the control performance against uncertainties is improved and chattering phenomena are reduced.