Adaptive Neural Network Tracking Control for Double-Pendulum Tower Crane Systems With Nonideal Inputs

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journalpeer-review

19 Scopus Citations
View graph of relations

Author(s)

Detail(s)

Original languageEnglish
Pages (from-to)2514-2530
Number of pages17
Journal / PublicationIEEE Transactions on Systems, Man, and Cybernetics: Systems
Volume52
Issue number4
Online published29 Jan 2021
Publication statusPublished - Apr 2022
Externally publishedYes

Abstract

A novel adaptive neural network tracking control method is systematically investigated for a unique double-pendulum tower crane system model in this article. Several critical and practical application-oriented control issues, including robustness, tracking error limitation, double-pendulum effects, and input dead zone nonlinearity, are considered simultaneously, which have never been well addressed in the existing literature. Technically, neural networks are employed to approximate the functions with uncertain/unknown dynamics and nonideal inputs. Several barrier Lyapunov functions are proposed that can circumvent the violation of tracking error limitations in the proposed control method. Importantly, based on the designed adaptive neural network tracking control method, the jib and trolley can track their desired trajectories very fast, and the hook and payload sway can be completely eliminated. The Lyapunov stability theory and Babalat's lemma are utilized to theoretically prove the convergence and stability of the proposed control system. Finally, well-designed simulation studies are carried out to verify the excellent performance and strong robustness of the control method. This article should be the first work considering a double-pendulum tower crane system with guaranteed convergence and performance without any linearization for the original nonlinear dynamic model.

Research Area(s)

  • Barrier Lyapunov function (BLF), dead zone, double-pendulum effects, neural network, radial basis function, tower cranes