Existing autonomous berthing technologies rely on precise mathematical ship models and mostly employ empirical formulas for modeling. However, in actual berthing scenarios, influenced by environmental factors and speed, these methods cannot accurately reflect the current ship maneuvering status in real-time, leading to limited berthing control accuracy. To address the aforementioned problems,

a ship autonomous berthing control method based on physics- informed neural networks (PINN) is proposed. The method constructs a real-time dataset using a sliding window and identifies ship maneuvering parameters in real-time through the physics-informed neural network. An adaptive controller based on gain scheduling is designed to dynamically adjust control gains using the identified parameters, realizing precise ship berthing.

Experimental results demonstrate that the PINN network can converge rapidly under dynamic conditions and accurately identify ship parameters, with a goodness of fit reaching 0.97. In berthing experiments, the method ensured that the terminal heading deviation and lateral error converged to a minimal range, achieving smooth and safe docking, with a heading error of 0.13°.

The method effectively resolves the failure of traditional control algorithms caused by model mismatch under unknown ship parameters and complex working conditions, offering a safe and interpretable adaptive berthing control solution.

To systematically review the technological evolution of unmanned surface vehicles (USVs) and explore the path of their convergence with intelligent ships, aiming to overcome the performance bottlenecks of individual USVs regarding endurance, computing power, and communication.

It reviews the centennial evolution of USVs, tracing the transition from radio remote control to fully autonomous navigation, and from single-agent operation to swarm collaboration. It provides an in-depth analysis of four core technologies: environmental perception, decision planning, motion control, and communication links. On this basis, the study focuses on the convergence trend between USVs and large intelligent ships, analyzing the "mothership-drone" cross-domain collaborative operational mode and the cloud-based management system driven by digital twins.

It indicates that current USV technology is undergoing an intelligent transition from "perception-avoidance" to "cognition-gaming". Furthermore, the "mothership-drone" collaborative mode, by combining the platform advantages of large ships with the high maneuverability of USVs, effectively resolves the challenges of individual USV operations in complex deep-sea environments and the "last mile" maneuvering difficulties for large intelligent ships entering and leaving ports, thereby achieving complementary advantages.

Collaborative mode represents a mainstream paradigm for future maritime operations. However, continuous breakthroughs are still required in areas such as regulatory adaptability, communication network security, and green energy propulsion. The findings provide theoretical references for constructing a new integrated air-surface-underwater intelligent maritime equipment system.

To review the current state of research on autonomous navigation decision-making and control technologies for intelligent unmanned surface vehicles, and to clarify the technical bottlenecks and development trends under scenarios of varying complexity,

a systematic investigation is conducted into the development history of key technologies for unmanned surface vehicles both domestically and internationally. It review addresses the differing technical requirements between low-to-medium complexity and high-complexity application scenarios, covering path planning, line-of-sight guidance, autonomous collision avoidance, automatic docking and undocking, multi-agent cooperative control, and autonomous recovery. It evaluates existing technological shortcomings and provides recommendations for future development.

Analysis indicates that autonomous navigation technology for open waters has matured and is gradually being implemented in engineering applications. However, core technologies for complex waters and complex missions still face developmental bottlenecks.

Looking further ahead, we propose establishing a standardized simulation and real-vessel testing evaluation system tailored to real-world scenarios. It will accelerate the rapid iteration and implementation of key technologies, thereby supporting the advancement of autonomous navigation decision-making and control technologies for unmanned surface vehicles in China.

In order to improve the ship in the actual berthing path control accuracy and other issues,

a ship berthing path planning method is proposed based on nonlinear model predictive control (NMPC) combined with moving horizon estimation (MHE), enabling future trajectory prediction and real-time control updates. A Fossen model of the ship is established in the four-degree-of-freedom (4-DOF): heave, pitch, yaw and roll. It adapts to the movement state and environmental changes of the ship on the water surface and improves the accuracy of berthing control. By simulating the simulation experiment of autonomous berthing path planning of papua new guinea and manila international port ships.

The results show that the trajectory error is less than 8.0 m and the berthing position error is only 0.6 m. The effectiveness, generalization and applicability of the proposed algorithm are verified. In order to more accurately simulate the ship's motion response in waves, a 4-DOF ship mathematical model is developed to provide more comprehensive ship motion information for the control system. It further realizes more accurate control and enhances the robustness of the system.

The method provides theoretical and practical support for autonomous berthing in diverse port environments.

To address the challenge of coordinated optimization between collision avoidance planning and motion constraints in the scenario where ships navigate close to dynamic surface targets, a hierarchical path planning method integrating the improved A^* algorithm, rapid reverse search iterative planning (RRSIP), and Hybrid A^* fine-grained planning, aiming to achieve efficient and kinematically compliant dynamic target tracking planning is proposed.

An improved A^* algorithm based on dynamic programming and oriented bounding box obstacle detection is used to plan a global reference path, reducing path length and redundant waypoints. For the dynamic target point, the RRSIP method is proposed, which reuses node information from prior searches via reverse search to quickly iterate and predict the approach point, avoiding global replanning and improving efficiency. Hybrid A^* algorithm is introduced near turning points for local refined planning, quickly generating a feasible path that satisfies ship kinematics and approach heading constraints.

Compared with other typical algorithms, the path length of the improved A^* algorithm proposed is reduced by an average of 4.17% and 1.79% respectively. The RRSIP method reduces the iterative planning time by at least 33.9% compared with the FR method. While ensuring path feasibility, the local Hybrid A^* planning reduces the time consumption by at least 72.1% compared with the global application.

The proposed method can effectively solve the problems of real-time performance and kinematic feasibility in dynamic target tracking, and significantly improve the autonomous tracking capability of ships in scenarios such as tugboat escort and maritime police law enforcement.

To improve the accuracy and robustness of ship trajectory prediction,

an ABiM-Ship network that encodes historical trajectories using a bidirectional selective state space model is proposed. An attention mechanism to explicitly align trajectories with heading and speed is utilized. A two-stage end-to-end joint prediction is designed, first regressing future trajectories, heading, and speed, then refining them using residual correction. Huber loss is introduced to constrain physical errors and stabilize convergence.

The experimental results show that this network outperforms traditional mainstream baselines in terms of average prediction error over short, medium, and long distances, achieving high prediction accuracy. The representation method, two-stage structure, and Huber loss all contribute significantly to performance gains.

The research findings achieve explicit coupling and coarse-to-fine prediction for trajectories, heading, and speed while maintaining linear temporal complexity. They have good reproducibility and scalability, providing a generalizable technical path and engineering reference for intelligent navigation and collaborative scheduling in complex maritime areas with high traffic density.

In order to improve the efficiency of tugboat operation scheduling and energy utilization,

an improved intelligent load forecasting algorithm integrating dynamic data preprocessing and online learning is proposed. Based on a hybrid LSTM-Adaboost architecture, the algorithm addresses the issues of temporal feature degradation and multimodal data fusion through the integration of differentiated LSTM weak predictors and dynamic weight allocation (error sensitivity penalty mechanism), and designs an online learning trigger mechanism (automatic retraining based on prediction error threshold) to achieve dynamic model updating. Additionally, an environmental data collaborative optimization module, including tidal information, is introduced to enhance the adaptability of load forecasting to port conditions. The algorithm is compared with conventional LSTM-Adaboost to validate its effectiveness.

The results indicate that after iterative optimization, the mean squared error of the improved algorithm is reduced by 40.8% compared to the conventional LSTM-Adaboost algorithm, demonstrating higher prediction accuracy and environmental adaptability.

The research results can provide a reference for tugboat energy optimization, safety management, and intelligent scheduling in ports.

To objectively and systematically understand the current status of reliability testing of maritime autonomous surface ships (MASS),

the current research status from three aspects: testing methods, testing technologies, and evaluation systems, and discusses the future development trends are analyzed. Specifically, it includes: conducting a visual analysis of 134 related papers using CiteSpace and VOSviewer to systematically sort out the research directions and development trends in the field of ship collision avoidance capability testing; sorting out the uses, advantages and disadvantages, and research status of the three major testing platforms: real ship testing, model testing, and virtual simulation testing; in-depth discussion on the development trends, feature comparisons, and challenges faced by the three mainstream testing scenario generation technologies based on expert knowledge, random sampling, and artificial intelligence; summarizing the evaluation indicators from four dimensions: data authenticity, scene complexity, risk, and generation efficiency; and on this basis, looking forward to future research directions.

The results show that virtual simulation testing has the advantages of low cost and high coverage and has become the main testing method. The ship collision avoidance capability testing method based on artificial intelligence has development potential in high-risk edge scenarios and ship interaction games, but the current research still faces challenges such as idealized motion models, lack of multi-ship dynamic game mechanisms, single evaluation indicators, and difficulties in virtual-to-real migration.

The research on testing scenario generation and deduction based on artificial intelligence has important research value and significance for promoting the testing of MASS.

To investigate the carbon fiber reinforced polymer (CFRP) hull lightweighting effect on the environmental impact,

Life cycle assessment on an 11 m CFRP high-speed vessel is conducted, focusing on atmospheric pollution indicators: global warming potential (GWP) and ozone depletion potential (ODP).

The results demonstrate that the fiber content adjustment-based lightweight design algorithm can achieve a 12.5% reduction of hull structure mass while maintaining structural safety by increasing fiber content from 40% (original case) to 55% (lightweight design case) approximately. However, due to the significantly higher environmental burden of carbon fiber production compared to resin, the manufacturing phase saw increases of 10.24% in GWP and 14.37% in ODP. Conversely, the operational phase benefited from reduced fuel consumption due to lightweighting, saving 323.98 t of fuel over 25 years, which decreased GWP and ODP by 4.13% and 4.19%, respectively.

The operational phase ultimately offset the negative environmental impacts of the manufacturing phase. Critical insights for green ship design and maritime industry decarbonization strategies is provided.

Aiming at the problem of the impact of center body position on nozzle cavitation intensity and flow morphology,

based on the CFD-Fluent, the Mixture multiphase model is employed, the k-ε turbulence model, and the Schnerr-Sauer cavitation model, numerical simulations are performed for nozzle flow fields with varying center body positions. The validity and reliability of the methodology are confirmed through comparison with prior research results.

The nonlinear regulatory effect of the center body's axial position on the internal flow field and cavitation intensity within the nozzle is systematically quantified. It clearly identified the junction of the nozzle throat and diffuser section as the optimal position most prone to triggering cavitation effects. Cavitation intensity and mass transfer rate peaked when the center body is located at this position. Cavitation is absent on the center body when positioned inside the nozzle. When center body located outside the nozzle, the downstream extent of the cavitation zone changed minimally, and cavitation intensity gradually diminished as the center body moved further downstream. Furthermore, based on the large eddy simulation (LES) method, an in-depth analyze the complex unsteady flow structures and large-scale radial diffusion characteristics of the vapor phase downstream of the nozzle under the optimal cavitation position condition is further conducted. Significant unsteady features are observed at a location 10 nozzle diameters (10D) downstream, where the vapor phase distribution expanded radially to four times the nozzle diameter (4D).

The research clarifies the regulatory mechanism of center body position on cavitation intensity, providing a theoretical basis for optimizing cavitating nozzle design. The findings contribute to enhancing the efficiency of industrial processes reliant on cavitation, such as cleaning and fragmentation, and offer theoretical support for developing adjustable center body structures to enable real-time control of cavitation intensity.