Despite receiving much attention from researchers in the field of naval architecture and marine engineering since the early stages of modern shipbuilding, the berthing phase is still one of the biggest challenges in ship maneuvering due to the potential risks involved. Many algorithms have been proposed to solve this problem. This paper proposes a new approach with a path-planning algorithm for automatic berthing tasks using deep reinforcement learning (RL) based on a maneuvering simulation. Unlike the conventional path-planning algorithm using the control theory or an advanced algorithm using deep learning, a state-of-the-art path-planning algorithm based on reinforcement learning automatically learns, explores, and optimizes the path for berthing performance through trial and error. The results of performing the twin delayed deep deterministic policy gradient (TD3) combined with the maneuvering simulation show that the approach can be used to propose a feasible and safe path for high-performing automatic berthing tasks.

Among the 6 degrees of freedom (6-DoF), excessive roll motion is the most dangerous cause of ships capsizing. However, when analyzing the maneuverability of surface ships, the roll components have usually been ignored. It is widely known that the influence of roll moment becomes significant for surface ships with low GM (metacentric height) and high speed. This paper examines the pure roll test for several surface ships to assess the roll-related hydrodynamic derivatives of added mass and damping in maneuvering. The objective ships are the KRISO Container Ship (KCS), David Taylor Model Basin (DTMB), Office of Naval Research Tumblehome (ONRT), and Delft 372 catamaran, where the DTMB and ONRT ships are equipped with complementary bilge keels as damping devices and have a small GM, which the Delft 372 catamaran does not have. The flow during pure roll is analyzed by the Computational Fluid Dynamics (CFD) simulation method that allows the complex flow around ships to be captured, especially when the bilge keel and skeg are considered. The results indicate that the roll moment is greatest in the catamaran. Since the roll moments of the DTMB and ONRT are larger than that of the KCS, bilge keels and surface shape also contribute to increasing roll damping moment. In addition, a comparison of the damping derivatives due to roll rate with results obtained from another method indicates that CFD simulation is capable of accurately predicting the roll-related derivatives, which is difficult to perform by the experiment method.

Hairy fairing cables are used extensively in towed missions because they reduce cable vibration, drag, lift forces, and acoustic noise. On the other hand, accurately estimating the hydrodynamic forces on each cable element of the hairy fairing cables poses a significant challenge in experimentation and computational fluid dynamics (CFD) because of the complex shapes with small hairs. This study examined the depth and tension of the towed cable system based on the proposed mathematical model with a hairy fairing cable. As mentioned in the mathematical model, the hydrodynamic forces of the cable significantly affected the towed cable posture. Therefore, the hydrodynamic forces of the hairy fairing cable were calculated using the CFD method, considering various modeling shapes with a range of inclined angles and Reynolds numbers. The simulation results of the cable depth were compared with sea trial data to validate the accuracy of the proposed approach and hairy shape modeling. This comparative analysis confirmed the reliability of the proposed approach in estimating the posture of the towed cable. In addition, the results also indicate a significant dependence of the hydrodynamic coefficient on the Reynolds number and the inclination angle of the cable element.

To reach a port, a ship must pass through a shallow water zone where seabed effects alter the hydrodynamics acting on the ship. This study examined the maneuvering characteristics of an autonomous surface ship at 3-DOF (Degree of freedom) motion in deep water and shallow water based on the in-port speed of 1.54 m/s. The CFD (Computational fluid dynamics) method was used as a specialized tool in naval hydrodynamics based on the RANS (Reynolds-averaged Navier-Stoke) solver for maneuvering prediction. A virtual captive model test in CFD with various constrained motions, such as static drift, circular motion, and combined circular motion with drift, was performed to determine the hydrodynamic forces and moments of the ship. In addition, a model test was performed in a square tank for a static drift test in deep water to verify the accuracy of the CFD method by comparing the hydrodynamic forces and moments. The results showed changes in hydrodynamic forces and moments in deep and shallow water, with the latter increasing dramatically in very shallow water. The velocity fields demonstrated an increasing change in velocity as water became shallower. The least-squares method was applied to obtain the hydrodynamic coefficients by distinguishing a linear and non-linear model of the hydrodynamic force models. The course stability, maneuverability, and collision avoidance ability were evaluated from the estimated hydrodynamic coefficients. The hydrodynamic characteristics showed that the course stability improved in extremely shallow water. The maneuverability was satisfied with IMO (2002) except for extremely shallow water, and collision avoidance ability was a good performance in deep and shallow water.