Xiao Lang

With the rapid advancement of artificial intelligence (AI), maritime autonomous surface ships (MASSs)—also referred to as autonomous vessels—have emerged as an important avenue of research in the maritime industry

1. Shipping currently accounts for 2.89% of global greenhouse gas (GHG) emissions, equivalent to about 940 million tons of CO₂ annually. Without targeted interventions, this proportion is expected ...

To address the challenges of safe and efficient automatic berthing for underactuated inland ships without tugboat or lateral thruster assistance, this article introduces a phased predictive berthing control (PPBC) method designed for low-speed conditions with uncertainties. A maneuvering modeling group (MMG) model is initially developed, incorporating twin-propeller and twin-rudder configurations, environmental disturbances (i.e., wind and currents), and model uncertainties. Three distinct berthing strategies, i.e., bow-in berthing, parallel berthing, and stern-in berthing, are formulated to ensure the applicability for most of berthing scenarios with phased states constraints. A time-delay observer is designed to estimate environmental disturbances and model uncertainties, on the basis of which a reliable predictive model is constructed by using the low-speed MMG model and the estimated disturbances and uncertainties. Leveraging this model, a model predictive control (MPC) method is designed to achieve the trajectory tracking of berthing control, while accounting for both control input and phased-state constraints. Simulation results demonstrate that the proposed PPBC method outperforms conventional proportional–integral–derivative, sliding mode control, and standard MPC methods in terms of trajectory, heading, and speed tracking error across a range of evaluated berthing scenarios.

This paper aims to enhance ship collision awareness by providing decision support in collision avoidance. To enable real-time computation in navigation, a Non-iterative Action Distance Determination Algorithm (NADDA) is developed. NADDA reduces the time complexity from ${O}\left( T \log n \right)$ to ${O}\left( T\right)$ by eliminating iterative processes, while maintaining the same space complexity as conventional bisection-based methods. This improvement significantly enhances computational efficiency, facilitating real-time applications. Based on NADDA, a robust and interpretable method for quantifying collision stages is developed by incorporating operational uncertainties, ship man-euverability, and the International Regulations for Preventing Collisions at Sea (COLREGs), thereby providing proactive colli-sion avoidance timing. Simulation experiments and a historical collision accident validated the effectiveness of the proposed methods. The results demonstrate that NADDA reduces computa-tion time by up to 93.1% compared to existing bisection-based methods, and the proposed collision stages quantification method achieves higher accuracy due to the consideration of operational uncertainties. The proposed method exhibits strong robustness and can provide real-time decision support in multi-ship encounter scenarios involving up to four target ships.