Observer-based H∞ fuzzy fault-tolerant switching control for ship course tracking with steering machine fault detection

Xuelin Zhang; Xiaobin Xu; Jianning Li; Yang Luo; Guodong Wang; Georg Brunauer; Schahram Dustdar

doi:10.1016/j.isatra.2023.05.021

Observer-based H_∞ fuzzy fault-tolerant switching control for ship course tracking with steering machine fault detection

ISA Trans. 2023 Sep:140:32-45. doi: 10.1016/j.isatra.2023.05.021. Epub 2023 May 31.

Authors

Xuelin Zhang¹, Xiaobin Xu¹, Jianning Li², Yang Luo³, Guodong Wang⁴, Georg Brunauer⁵, Schahram Dustdar⁶

Affiliations

¹ China-Austria Belt and Road Joint Laboratory on Artificial Intelligence and Advanced Manufacturing, Hangzhou Dianzi University, Hangzhou 310018, China; School of Automation, Hangzhou Dianzi University, Hangzhou 310018, Zhejiang, China.
² School of Automation, Hangzhou Dianzi University, Hangzhou 310018, Zhejiang, China. Electronic address: ljn@hdu.edu.cn.
³ Hangzhou Qianhang Shipyard Co., LTD, Hangzhou 311256, Zhejiang, China.
⁴ Sino-Austria Research Institute for Intelligent Industries (Nanjing) Co.,Ltd., Nanjing 210003, China.
⁵ TU Wien, Institute for Energy Systems and Thermodynamics, Getreidemarkt 9, 1060 Vienna, Austria.
⁶ Distributed Systems Group, TU Wien, 1040 Vienna, Austria.

PMID: 37295998
DOI: 10.1016/j.isatra.2023.05.021

Abstract

To enhance the robustness of ship autopilot (SA) system with nonlinear dynamics, unmeasured states, and unknown steering machine fault, an observer-based H_∞ fuzzy fault-tolerant switching control for ship course tracking is proposed. Firstly, a global Takagi-Sugeno (T-S) fuzzy nonlinear ship autopilot (NSA) is developed with full consideration of ship steering characteristics. And the actual navigation data collected from a real ship are used to verify the reasonableness and feasibility of NSA model. Then, virtual fuzzy observers (VFOs) for both fault-free and faulty systems are proposed to estimate the unmeasured states and unknown fault simultaneously, and compensate for the faulty system by using the fault estimates. Accordingly, the VFO-based H_∞ robust controller (VFO-HRC) and fault-tolerant controller (VFO-HFTC) are designed. Subsequently, a smoothed Z-score-based fault detection and alarm (FDA) is developed to provide switching signals for which the controller and its corresponding observer should be invoked. Finally, simulation results on the "Yulong" ship demonstrate the effectiveness of the developed control method.

Keywords: Fault detection and alarm; Fault-tolerant control; Smoothed Z-score algorithm; Steering machine fault; Takagi–Sugeno fuzzy nonlinear ship autopilot; Virtual fuzzy observer.