ROV recovery with wave-motion compensation using model predictive control

Abstract

Recovery of ROVs with a moonpool-based launch-and-recovery system on a small surface vessel is studied. We assume the ROV recovery takes place within the moonpool, where the negatively buoyant ROV/latchbeam assembly is influenced by the forces induced by the surface waves as well as hydrodynamic interaction forces from the surface vessel hull in addition to the wires and thrusters. We study a model predictive controller (MPC) that internally predicts the ROV’s pendulum-like motion and wire tension that is driven by the surface vessel motions, which is assumed to be measured. The main contribution in the paper is the use of a simple cart–pendulum model to approximate the wire forces by modeling the wires are rigid rods. This is verified to achieve low computational complexity to enable real-time MPC computations. The performance of this combined feedback and feed-forward wave compensation strategy is compared to conventional PID and LQR controllers for ROV motion. Simulations use rigorous models of the multi-body dynamics, including hydrodynamics and wire elasticity. The simulation results show improvements in operability in terms of reduced relative motion between the two bodies, avoidance of negative tension, and thereby the potential of less risk for wire snap and ROV impact within the surface vessel’s moonpool.