DDPG-based model adaptive cooperative control for path following of a multi-vessel towing system

Conventional control methods often struggle to achieve both precise control and accurate parameter estimation for strongly coupled maritime towing systems under model uncertainty and environmental disturbances. To overcome this, we propose a coordinated method that integrates online model adaptation into a Model Predictive Control (MPC) framework. The adaptively compensated model of the towed vessel is embedded within an MPC-based supervisor, allowing accurate prediction of the vessel’s states and effective allocation of tugboat control commands. Simulation results show that the proposed method maintains high-precision path- following performance both in calm conditions and under external disturbances, thereby demonstrating its effectiveness and robustness in handling model uncertainties and environmental disturbances.