Multi-AUV cooperative hunting requires autonomous underwater vehicles (AUVs) to strategize the encirclement of evaders while navigating around obstacles and other AUVs. Despite the promise of multi-agent reinforcement learning (MARL) in continuous control problems, its low sample efficiency poses a challenge in unknown environments and complex control scenarios. To overcome these limitations, we present a Knowledge-Guided Reinforcement Learning (KG-RL) approach, which integrates an Artificial Potential Field (APF) to enhance sample efficiency and operational safety. Our methodology is bifurcated into pre-training and fine-tuning phases. During the pre-training phase, an APF is employed to generate a concise set of demonstration trajectories that provide agents with foundational knowledge. Subsequently, the fine-tuning phase leverages real-time APF knowledge to direct the learning process, encouraging agents to balance following demonstrated actions with seeking out more optimal solutions. We assess the efficacy of our method through extensive simulations across diverse tasks, demonstrating its ability to expedite the learning process and yield more strategic decision-making. Our approach achieves superior results compared to traditional MARL benchmarks, particularly in learning efficiency, decision quality, and overall performance.

It is expected that the prototypes of unmanned merchant ships will be deployed in the next few years. However, there is no specific research on whether the introduction of unmanned ships will reduce the risk of ship collision accidents in which communication between vessels is critical. This work constitutes an attempt to bridge the gap identified above by applying the Hybrid Causal Logic (HCL) methodology to model general-level collision scenarios of unmanned ships. The HCL methodology has been selected for its proven applicability to risk assessments, even when empirical data may be insufficient. Collision scenarios involving unmanned ships have been created in which manned ships of the conventional collision scenario HCL model are replaced with unmanned ships. Then, collision scenarios capturing the interactions between a manned ship and an unmanned ship were modeled. By comparing the qualitative and quantitative results of the different scenarios, we can see that the introduction of unmanned ships may effectively reduce the occurrence of ship collision accidents.

Collision avoidance algorithms play a crucial role in ensuring the safety and effectiveness of autonomous ships, which require comprehensive testing in realistic multi-ship encounter scenarios. However, existing scenario generation methods often inadequately represent the spatiotemporal complexity and dynamic risk interactions of real-world encounters, leading to biased evaluations. To bridge this gap, this paper proposes a combinatorial-testing-based scenario generation framework integrated with spatiotemporal complexity optimisation. First, a full-process scenario representation model is developed by abstracting real-world navigation features into a discretised parameter space. Subsequently, a combinatorial-testing-based scenario generation method is adopted to cover the parameter space, generating a high-coverage scenario set. Finally, spatiotemporal complexity is introduced to filter out oversimplified scenarios and extremely dangerous scenarios. Experiments demonstrated that 13.7% of generated scenarios were eliminated as unrealistic or trivial, while high-risk encounter scenarios and multi-ship interaction scenarios were amplified by 7.96 times and 5.84 times, respectively. Compared to conventional methods, the optimised scenario set exhibited superior alignment with real-world complexity, including dynamic risk escalation and multi-ship coordination challenges. The proposed framework not only advances scenario generation methodology through its integration of combinatorial testing and complexity-driven optimisation, but also provides a practical tool for rigorously validating autonomous ship safety systems.

A novel scheme is proposed for the distributed multi‐ship collision avoidance (CA) problem with consideration of the autonomous, dynamic nature of the real circumstance. All the ships in the envisioned scenarios can share their decisions or intentions through route exchange, allowing them to make subsequent decisions based on the route planning in each iteration. By leveraging route exchange, the multi‐ship CA problem involves iterations for negotiation, and is regarded as a staged cooperative game under conditions of complete information. The concept of closest spatio‐temporal distance (CSTD) is introduced to more accurately assess collision risk between ships. A coordinated CA mechanism is established when a collision risk is identified, which further incorporates considerations including the stand‐on/give‐way relationships, negotiation rounds, route re‐planning calculation, as well as the cost factor for route evaluation. The Nash bargaining solution (NBS) is elaborated to achieve Pareto‐optimal CA routes in the scenarios. In the proposed model, while the individual interest of each ship are maximized, the economic fairness and global optimization of the overall system are also maintained. Simulation results indicate that the NBS shows good flexibility and adaptability, and that when all ships comply with route re‐planning solution, the proposed scheme can bring out normal solutions within a limited number of re‐planning iterations.

Ship collision avoidance (CA) is the most fundamental and essential issue of maritime safety, regardless of whether the ship is a conventional ship or a future maritime autonomous surface ship (MASS). However, effective ship-to-ship communication has been recognized as a critical issue and one of the major challenges affecting the success of ship CA. This study intends to investigate the basic trust concerns in the cybersecurity components of the ship CA process and develop a blockchain-based solution for trustworthy communication. The novel idea behind applying blockchain to ship CA decision-making is that the various participating ships in a scenario of dynamically changing ship encounters form a decentralized network of opportunities, which makes blockchain an appealing tool to offer a solution for evaluating and maximizing trust in entity dynamics. The mechanism of entity participation in upholding a master chain for avoiding ship encounters is described in this work. The research begins by analyzing conventional ship CA decision scenarios under COLREGs in order to identify formal communication strategies that can make information more understandable. Based on this, it is demonstrated how beliefs of trust (BoT) between entities are encoded and combined on the chain to enable entities in the encounter scenario to form an initial opinion about another entity before becoming familiar with it. This first encounter occurs during ship CA. Second, this article examines how encounter ships have temporary rights in the process of making CA decisions. This generates blocks and attaches them to the chain at the consensus layer of blockchain technology. Finally, this study undertakes a case test of ship CA based on the aforementioned communication principles. In order to ensure safe and effective communication during brief, one-time ship encounters, the results offer a practical method for conveying CA reasoning in encounter scenarios. As a result, it can aid in the selection of better (more reliable) nodes to uphold the evidence-based understanding of the ship CA process and assess the reliability of unidentified members.