Intelligent-Feedback Hybrid Control Strategy for 5-Dof Ship-Mounted Rotary Cranes in Consideration of Safety

Abstract

The autonomous anti-sway control of the 5D ship-mounted crane system is a tough issue due to the under-actuated characteristic, strong coupling, base excitation, and unknown system parameter. Furthermore, for security considerations, the under-actuated states (i.e., the swing angles) are required to be limited in certain ranges, which make the control issue more difficult. In this paper, a hybrid anti-sway control method which combines the radical basis neural network (RBFNN), the genetic algorithm (GA) and the traditional feedback control algorithm is proposed. Specially, the RBFNN is used to estimate the unknown dynamics online. On basis of this, a parameter independent feedback control method is designed to regulate the state variables and constraint the swing angles. The GA is used to optimize the control gains and the center and the width of the hidden neuron. The closed loop stability is analyzed and the effectiveness of the control method is validated via simulation results.

Keywords: Genetic algorithm, intelligent control, neural network, ship-mounted crane, under-actuated system