Anti-sway Control for Overhead Cranes Using Neural Networks

This paper presents an anti-sway control for overhead cranes using neural networks. To reduce the sway of a load after positioning to the greatest extent possible, we construct a trajectory for the position of the trolley. Radial basis function networks (RBFNs) are employed to generate the desired trolley position. Thereafter, a particle swarm optimization (PSO) is used as a learning algorithm in which the maximum swing angle after positioning is adopted as the objective function to be optimized. By moving the trolley along the trajectory thus generated, the sway angle can be suppressed. The performance of the proposed anti-sway control is confirmed by numerical simulations. Furthermore, the realization and effectiveness of the present approach are verified by experiment.