Uncertainty-aware ship trajectory prediction via Spatio-Temporal Graph Transformer
Jincheng Gong, Huanhuan Li, Hang Jiao, Zaili Yang

Project |

• Introduced a novel ship trajectory prediction benchmark for Maritime Autonomous Surface Ships (MASS).
• Developed an advanced trajectory prediction algorithm that integrates spatio-temporal and probabilistic modelling.
• Pioneered an uncertainty-aware feature modelling approach to improve prediction reliability and interpretability.
• Validated the model’s performance across three benchmark datasets using three key evaluation metrics.

LLM4STP: A large language model-driven multi-feature fusion method for ship trajectory prediction
Hang Jiao, Jincheng Gong, Huanhuan Li, Jasmine Siu Lee Lam, Yaqing Shu, Jin Wang, Zaili Yang

Project |

• Propose LLM4STP, a novel multi-feature fusion method integrating LLMs and maritime knowledge for improved ship trajectory prediction.
• Introduce an adaptive graph-masked Transformer for dynamic ship interaction modelling, capturing complex spatial relationships.
• Combine local convolution and global LLM inference for effective multi-scale temporal modelling.
• Establish a system for trajectory uncertainty quantification using 2D Gaussian heatmaps and Geohash-based encoding.

Cybersecurity in smart port systems: A systematic review and data-driven research agenda
Yunfeng Zhao, Jincheng Gong, Huanhuan Li, Trung Thanh Nguyen, Christian Matthews, Zaili Yang

• Examines attack scenarios, technologies, and policy evolution in smart port cybersecurity.
• Reviews smart port cybersecurity literature to identify key research clusters and emerging themes.
• Utilises empirical data from the Maritime Cyber Attack Database (MCAD) to construct a global cyber-attack map.
• Identifies six research challenges, including CPS vulnerabilities, human factors, and AI-related threats.

Data Clustering and Machine Learning in the Internet of Marine Things
Huanhuan Li, Jincheng Gong, Zaili Yang

• This chapter reviews IoMT data clustering and ML methods with CiteSpace-based mapping of research trends.
• It showcases these methods in maritime safety, routing, port operations, and environmental monitoring via case studies.
• Key challenges addressed include data quality, scalability, and cybersecurity.
• Future directions focus on AI, digital twins, LLMs, and multimodal Transformers for autonomous maritime systems.

Applications of ship digital twins in maritime transport: A systematic review
Huanhuan Li, Jincheng Gong, Alan J. Murphy, Zaili Yang

• Develops a two-stage analytical framework for ship digital twin research.
• Maps the evolution of ship digital twins through scientometric analysis.
• Evaluates ship digital twin studies from modelling, system, and validation perspectives.
• Clarifies the distinction between digital shadows and bidirectional digital twins.
• Identifies deployment challenges and future directions for ship digital twins.

MC-QRA: Quantitative Spatiotemporal Risk Assessment and Probabilistic Dependency Analysis of Maritime Cyber Attacks
Jincheng Gong, Yunfeng Zhao, Huanhuan Li, Zaili Yang

• A quantitative framework integrates scenario-based and spatiotemporal maritime cyber risk assessment.
• Empirical cyber incident records support interpretable risk classification and quantification.
• High-risk attack scenarios and spatiotemporal patterns of cyber incidents are identified.
• Bayesian modelling captures dependencies among risk-influencing factors.
• Dependency-aware inference supports the assessment of high-risk maritime cyber scenarios.

A data-driven framework for spatiotemporal pattern mining of cyber incidents in digital maritime transport systems
Huanhuan Li, Jincheng Gong, Yunfeng Zhao, Zaili Yang

• Developed a data-driven framework for spatiotemporal maritime cyber risk analysis.
• Constructed MCAD++ to enable structured analysis of maritime cyber incidents.
• Examined cyber threat associations and temporal relationships using association rules and transfer entropy.
• Explored spatial patterns of maritime cyber incidents using HDBSCAN and Moran’s I.