Dynamic Positioning Using Model Predictive Control With Short-Term Wave Prediction
Henning Øveraas, Håkon S. Halvorsen, Olav Landstad, Vidar Smines, Tor Arne Johansen
Abstract
Remotely operated vehicle (ROV) operations are today typically supported by large designated vessels. New emerging concepts aim to streamline ROV operations by utilizing unmanned surface vessels of a smaller size. Reduction in size may result in first-order wave induced motion being more significant. This motivates the use of dynamic positioning control using thrusters to actively compensate for first-order wave-driven horizontal-plane motion in order to maximize operability. This article proposes a controller for dynamic positioning based on model predictive control and short-term wave motion prediction intended to actively compensate for first-order waves. By considering the full dynamic sea environment, the controller is able to dampen out some of the oscillatory motion caused by first-order waves. The controller is able reduce the average deviation from the set-point with up to 65% for a variety of sea conditions. The maximum distance error to the reference point is reduced by up to 65% depending on sea state. The dynamics of the thrusters is a limiting factor when counteracting first-order waves and fast thrusters are therefore crucial in achieving best possible positioning. The cost of the wave-compensated positioning is a more dynamic consumption of power.