This paper proposes a multi-agent control scheme for multiple physically interconnected tugboats performing a ship towing process. These tugs are coordinated by two control layers. In the higher layer, the supervisory controller outputs the desired towing forces and reference trajectories for the tugs. This information is used by a tug’s local controller in the lower layer to calculate the thruster forces and moment for manipulating the ship. The control strategy is based on the model predictive control concept, with the performance function designed by using the position and velocity error to make the ship follow waypoints and speed profile. A distributed control architecture is designed based on the alternating direction method of multipliers (ADMM) to reach a consensus between the predicted position generated by the tug controllers and the reference trajectory generated by the supervisory controller. Simulation experiments illustrate that the proposed method ensures the smooth and efficient maneuverability of the towing process.
展开▼
