Adaptive steering control for uncertain ship dynamics and stability analysis

  • Authors:
  • Nazli E. Kahveci;Petros A. Ioannou

  • Affiliations:
  • AVL Research and Engineering, TUBITAK MAM Teknoloji Serbest Bolgesi, B Blok 207-208, Gebze, 41470, Turkey;University of Southern California, 3740 McClintock Avenue, EEB 200B Los Angeles, CA 90089-2562, United States

  • Venue:
  • Automatica (Journal of IFAC)
  • Year:
  • 2013

Quantified Score

Hi-index 22.14

Visualization

Abstract

Marine transportation offers a cost-effective and viable alternative to cargo aircraft, trains, and trucks in general. Traffic control in congested waterways has recently become a challenging task demanding higher accuracy in ship navigation and advanced methods for collision or grounding avoidance. Modern ship control systems call for new technologies to be applied to ship steering which involves station-keeping and course-changing maneuvers. Uncertain cargo ship dynamics require robust steering control design techniques potentially reducing the effects of external conditions and minimizing path deviations in case particularly strong lateral wind or wave forces are experienced. Among additional issues to be addressed are the possible conflicts between input limitations and controller performance. We present an adaptive steering control design for uncertain ship dynamics subject to input constraints while avoiding performance compromises under changing environmental conditions. An adaptive law is combined with a control design including a Linear Quadratic (LQ) controller and a Riccati based anti-windup compensator using Certainty Equivalence Principle for asymptotically stable plants with saturation limits imposed on the control input. If the desired linear performance can be recovered for nominal systems subject to input saturation by implementing an LQ control augmented with an anti-windup compensator, our analysis on its adaptive counterpart reveals that the perturbation terms due to plant parameter errors in the adaptive scheme do not cause any unbounded signals in the closed-loop, and the system remains stable. Continuous-time lumped Hurwitz systems with input saturation nonlinearities and unknown plant parameters can benefit from the resulting indirect adaptive control design. Several ship maneuvering scenarios are simulated to verify the effectiveness of our approach.