Cooperative Control of Autonomous Multi-Vessel Systems for Floating Object Manipulation

This thesis provides a set of cooperative control schemes for autonomous multi-vessel systems to manipulate a floating object through physical interconnections in onshore (inland waterways and ports) and offshore areas. Thanks to the maturity and popularity of the advancing technologies in information, communication, sensors, automatic control, and computational intelligence, we have seen the application scenarios of the autonomous vessels being gradually extended from fundamental research to civil and commercial uses. In recent years, to ensure that the regulatory framework for autonomous vessels keeps pace with technological developments, the International Maritime Organization (IMO) has started to include the autonomous vessels issue in its sessions. Maritime operations have become more complex and their scale is getting larger, requiring the involvement of multi-vessel systems. In recent decades, the formation control of multiple vessels has been investigated and several mature control methods are proposed to cope with different typical missions. However, there is a lack of research focusing on floating object manipulation by multiple autonomous vessels through physical interconnections. Thus, the research question of this thesis is How to design a scalable control scheme for multiple ASVs to manipulate a floating object through physical interconnections? Through the analysis of four typical manipulation ways in the maritime field, the towing way is selected in this thesis as the basic physical manipulation model, which has advantages in good maneuverability of the floating object, better safety of the manipulation system, and more flexibility of the operational scenarios. The dynamic model of the towing system is built by using the 3 DOF vectorial representation, where the towing forces and towing angles are the kinetic and kinematic interconnections between the floating object and tugboats, respectively. Considering the multiple control inputs, multiple control constraints, and limited maneuverability of the towing system, the model predictive control (MPC) strategy is the research approach in this thesis. Furthermore, to achieve the distributed control architecture, the alternating direction method of multipliers (ADMM) is used. In this thesis, the proposed method is used in three different operational environments, port areas, inland waterways, and open sea for floating object manipulation...