Development of a Dynamic Positioning System for Merlin WR 200 ROV - Development and implementation of a model based control law for station keeping and path following

Abstract

Control of Remotely Operated Vehicles (ROVs) with manipulator arms require control of all degrees of freedom. In particular, it is necessary to be able to keep the vehicle at a constant position and orientation in order for the manipulator arms to efficiently perform its task. To increase the level of autonomy of a ROV, station keeping, trajectory tracking and path following can be implemented. These functions all require a controller and the development of such a controller is the key task in this thesis. The design of a path following algorithm for backtracking the path driven by the ROV when controlled by the operator is also one of the goals.

When designing a control system for a marine craft, the choice of controller is not obvious. Several controllers exist with different properties and demands. A desired feature of a controller is that it is robust to parameter uncertainty, since the parameters of a marine craft may change with different operating conditions. For an underwater vehicle, the ocean current is present as a disturbance, so the controller needs to be able to compensate this disturbance. In this thesis a position and velocity controller that adapts the system parameters and current disturbance is developed. Convergence of the position error and virtual velocity error to zero is proven through Lyapunov theory and by utilizing Barb\v{a}lat's lemma.

The developed controller has been implemented and tested on a Simulink simulation model of the Merlin WR200 ROV from IKM Subsea. The tests considered station keeping, trajectory tracking, path following of paths generated by the ROV as it drives through the ocean space and of predefined paths. For all cases, the ROV is able to converge to the desired position, velocity or path. The performance of the developed controller and 2 other controllers is compared for station keeping and trajectory tracking and the developed controller shows very good results. The controller was also tested for station keeping and trajectory tracking in the advanced simulator from CM Labs at the IKM headquarters in Bryne. The performance was very good under the conditions of the simulation.

Future work includes among other things developing a more extensive dynamic model for Merlin WR200 that includes the effects of cable drag, developing a steering law for 6 DOF underwater vehicles and conducting a sea trial.

The goal of this thesis was to develop a new model based control law for ROVs and verify its performance through simulations. This has been successfully accomplished.