Title: Dynamic neural-fuzzified adaptive control of ship course with parametric modelling uncertainties

Authors: Yang Wang, Chen Guo, Fuchun Sun, Zhipeng Shen, Di Guo

Addresses: College of Information Science and Technology, Dalian Maritime University, Dalian, 116026, China. ' College of Information Science and Technology, Dalian Maritime University, Dalian, 116026, China. ' State Key Laboratory of Intelligent Technology and Systems, Tsinghua University, Beijing, 100084, China; College of Information Science and Technology, Dalian Maritime University, Dalian, 116026, China. ' College of Information Science and Technology, Dalian Maritime University, Dalian, 116026, China. ' School of Instrument Science and Opto-electronics Engineering, Beijing University of Aeronautics & Astronautics, Beijing, 100191, China

Abstract: A dynamic neural-fuzzified system (DNFS) based adaptive control algorithm is presented in this paper, aiming at the uncertainties arising from changes of the model parameters in ship course control. The inverse dynamics of the Norrbin non-linear NARX ship model is identified by the DNFS properly, and the structure and parameters of the DNFS are adjusted simultaneously. Well-trained DNFS is then connected in parallel with a PD controller to construct an adaptive controller for course control, while the weights of DNFS are more adjusted by the adaptive law whose independent variable is the output of PD controller. Simulation results of the course tracking of a 5,446 TEU container ship validate the effectiveness of the proposed algorithm.

Keywords: dynamic neural-fuzzified system; DNFS; uncertainties; course tracking; adaptive control; neural networks; ship course; ship navigation; simulation; container ships; parametric modelling.

DOI: 10.1504/IJMIC.2011.041780

International Journal of Modelling, Identification and Control, 2011 Vol.13 No.4, pp.251 - 258

Published online: 21 Mar 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article