The causes of maritime accidents are complex, mostly due to the coupling of four types of factors: human-ship-environmental-management. To effectively analyze the causes of maritime accidents in China, and reveal the risk coupling characteristics of accidents, this paper establishes the N-K model of maritime accident, and calculates and analyzes the four types of coupling of risk factors affecting safety in maritime traffic. This paper collects 922 maritime accidents that occurred in China from 2000 to 2020, and analyzes the location, type, and level of accidents and uses the trigger principle to describe the process of accidents. For marine and inland river accidents, this paper calculates the four types of coupling values of risk factors (single-factor coupling, two-factor coupling, three-factor coupling, four-factor coupling) for comparison and analysis. In addition, this paper calculates the coupling values of six typical maritime accidents of collision, sinking, contact, fire/explosion, stranding, grounding. According to the coupling values and the frequency of sub-factors, this paper analyzes the coupling characteristics of maritime accidents. The results show that in maritime accidents, as the number of risk factors participating in the coupling increases, the coupling value increases, and the multi-factor coupling is more likely to cause accidents. The overall situation of risk coupling causes of maritime accidents is basically consistent with inland river accidents, but they have their own characteristics in the specific degree of risk coupling and the dominant risk elements. In different types of maritime accidents, the risk coupling has different characteristics, and the dominant risk factors are also different.

Maritime transport plays a critical role in global logistics. Compared to road transport, the pace of research and development is much slower for maritime transport. It faces many major challenges, such as busy ports, long journeys, significant accidents, and greenhouse gas emissions. The problems have been exacerbated by recent regional conflicts and increasing international shipping demands. Maritime Autonomous Surface Ships (MASSs) are widely regarded as a promising solution to addressing maritime transport problems with improved safety and efficiency. With advanced sensing and path-planning technologies, MASSs can autonomously understand environments and navigate without human intervention. However, the complex traffic and water conditions and the corner cases are large barriers in the way of MASSs being practically deployed. In this paper, to address the above issues, we investigated the application of Large Language Models (LLMs), which have demonstrated strong generalization abilities. Given the substantial computational demands of LLMs, we propose a framework for LLM-assisted navigation in connected MASSs. In this framework, LLMs are deployed onshore or in remote clouds, to facilitate navigation and provide guidance services for MASSs. Additionally, certain large oceangoing vessels can deploy LLMs locally, to obtain real-time navigation recommendations. To the best of our knowledge, this is the first attempt to apply LLMs to assist with ship navigation. Specifically, MASSs transmit assistance requests to LLMs, which then process these requests and return assistance guidance. A crucial aspect, which has not been investigated in the literature, of this safety-critical LLM-assisted guidance system is the knowledge and safety performance of the LLMs, in regard to ship handling, navigation rules, and skills. To assess LLMs’ knowledge of navigation rules and their qualifications for navigation assistance systems, we designed and conducted navigation theory tests for LLMs, which consisted of more than 1500 multiple-choice questions. These questions were similar to the official theory exams that are used to award the Officer Of the Watch (OOW) certificate based on the Standards of Training, Certification, and Watchkeeping (STCW) for Seafarers. A wide range of LLMs were tested, which included commercial ones from OpenAI and Baidu and an open-source one called ChatGLM, from Tsinghua. Our experimental results indicated that among all the tested LLMs, only GPT-4o passed the tests, with an accuracy of 86%. This suggests that, while the current LLMs possess significant potential in regard to navigation and guidance systems for connected MASSs, further improvements are needed.

It is crucial to develop a COLREGs-compliant intelligent collision avoidance system for the safety of unmanned ships during navigation. This paper proposes a collision avoidance decision approach based on the deep reinforcement learning method. A modified collision avoidance framework is developed that takes into consideration the characteristics of different encounter scenarios. Hierarchical reward functions are established to assign reward values to constrain the behavior of the agent. The collision avoidance actions of the agent under different encounter situations are evaluated on the basis of the COLREGs to ensure ship safety and compliance during navigation. The deep Q network algorithm is introduced to train the proposed collision avoidance decision framework, while various simulation experiments are performed to validate the developed collision avoidance model. Results indicate that the proposed method can effectively perform tasks that help ships avoid collisions in different encounter scenarios. The proposed approach is a novel attempt for intelligent collision avoidance decisions of unmanned ships.

Device-free indoor localization may play a critical role in improving passengers’ safety in large vessels, particularly for scenarios without equipped radios. However, due to dynamic internal and external influences from the sailing ship such as changing sailing speed, existing localization systems suffer huge accuracy degradation in a mobile ship environment. The challenges are mainly due to rich and arbitrary ship motions and the resulting complicated impacts on the indoor wireless channels. To address the challenges, in this paper, we first propose a ship motion descriptor to extract discriminative latent representation from complex ship motions by leveraging deep-learning techniques. Based on this representation, we then design a novel fingerprint roaming model, i.e., MoLoc, to automatically learn the predictive fingerprint variation pattern and transfer the online fingerprint measurement to adapt to dynamic ship motions in real-time. Furthermore, an unsupervised learning strategy is proposed to train the fingerprint roaming model using unlabelled onboard collected data which does not incur any labor costs. We have implemented and extensively evaluated MoLoc on real-world cruise ships, where experimental results demonstrate that MoLoc improves localization accuracy from 63.2% to 92.8% compared to the state-of-the-art localization methods including Pilot, LiFS, SpotFi, and AutoFi, while achieving a mean error of 0.68m.

Accurately interpreting regional traffic situations plays a pivotal role in emerging intelligent transportation systems, particularly in the evaluation of traffic states to realize the implementation of rational interventions. Nonetheless, existing studies face challenges when it comes to unveiling the complex nested interactions among multiple ships while simultaneously factoring in various influential elements for precise collision risk evaluation. This paper aims to develop a comprehensive methodology for collectively modeling, evaluating, and mitigating maritime traffic complexity, to enhance the comprehension of traffic patterns and guide anti-collision management in complex waters. First, a novel ship domain-based approach is proposed, incorporating individual ship attributes, relative bearing, ship motion dynamics, and restricted water geography to realize the accurate evaluation of ship-pair conflict risk. Subsequently, advanced motif structure-based indicators and a network disintegration model are merged to provide a thorough and nuanced characterization of the topological dependencies among multiple conflicts within a specified maritime region. Simultaneously, a comprehensive complexity evaluation approach, combining Principal Component Analysis (PCA) and a Fuzzy Clustering Iterative (FCI) method, is employed to achieve dependable parameterization and classification of traffic complexity levels. Finally, the collective impact of multiple interdependent conflicts on overall traffic complexity mitigation is investigated to support the identification of key influential conflicts that should take precedence in joint resolution efforts. Extensive experimental analyses based on Automatic Identification System (AIS) data are carried out to validate the effectiveness of the proposed methodology. These analyses demonstrate its applicability in accurately assessing conflict risk, hierarchically categorizing traffic complexity levels, and providing guidance for joint conflict resolution endeavors. Consequently, this methodology holds significant promise for bolstering the growth of intelligent transportation service systems and facilitating the automation of maritime traffic management.