Back to search

University of Sheffield

Nonlinear Model Predictive Control of Autonomous Surface Vehicle in Rough Seas

Abstract

dc:description.abstract

Autonomous Surface Vehicles (ASVs) offer the potential of performing dirty, dull, and dangerous missions at sea, in an automated fashion. However, without consideration of the ocean environment and the risk of wave-induced damage, the use of conventional ASVs is restricted to relatively calm sea-states. Assuming a conventional underactuated ASV with only throttle and rudder inputs, this thesis addresses the problem of controlling an ASV in an optimal fashion, while maintaining headway towards a desired destination in any sea and with any sea state. This is a challenging problem owing to the coupled, nonlinear nature of the vessel-wave dynamics, along with multiple competing performance objectives, such as excessive motions, and actuation effort. The thesis shows that the optimal solution, obtained via nonlinear model predictive control (NMPC), involves tacking at two different timescales. This subsequently motivates the design of a two-degree of freedom controller, consisting of a tacking planner that generates a long-term, optimal, heading and velocity reference, and a feedback regulator that produce the optimal throttle and rudder commands to maintain this reference, whilst minimizing wave-induced effects with smaller tacks. This thesis represents the first work to formulate and solve the optimal control problem of navigating in rough seas, based upon a coupled dynamic model of a 6-DOF vessel excited by waves. Closed-loop simulation results from a high fidelity ocean-vessel model demonstrate significant reductions in excessive vessel motions and forces, compared to a conventional path following PID controller for head and beam seas. Further examinations show the superiority of the NMPC in a full sea state. The impact of the prediction horizon parameters on the performance is investigated with a view towards an adaptive prediction horizon. Lastly, the ability of the NMPC to use one set of Response Amplitude Operators (RAOs) in other wave conditions without loss of performance is shown.

Degree

thesis:*
Name dc:type.qualificationname
Ph.D
Level dc:type.qualificationlevel
doctoral
Grantor dc:publisher.institution
University of Sheffield
Year dc:date.issued
2021

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • McCullough, Daniel Roger
Advisor dc:contributor.advisor
  • Rossiter, Anthony
ORCID person:identifier.orcid

Rights

Language dc:language.dcterms_iso639_2
eng

Identifiers

dc:identifier.*
Identifier
uk.bl.ethos.855690

Chain of custody

source
Harvested from
White Rose University Consortium
Base URL
etheses.whiterose.ac.uk/cgi/oai2
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
related terms
citation

McCullough, Daniel Roger. Nonlinear Model Predictive Control of Autonomous Surface Vehicle in Rough Seas. doctoral thesis, University of Sheffield, 2021.