(03)

Science

↓ (W31 ‘26 · Jul 27–Aug 2)
01openalex2026-07-29Si Hyun Kim, Tae Jung ParkDOI ↗ (opens in a new tab)

Defining High‐Impact <scp>AI</scp> in Autonomous Shipping: Legal Frameworks, Supply Chain Resilience, and Maritime Safety Dimensions (opens in a new tab)

The paper examines regulatory gaps in classifying high-impact AI for Maritime Autonomous Surface Ships (MASS), noting that the EU’s AI Act (2024) and South Korea’s AI Framework Act (2025) lack specific criteria for autonomous shipping. It proposes five criteria—navigation centrality, reliability under stress, systemic harm potential, data integrity dependence, and lack of human override—to define high-impact AI in this context. The authors link accurate classification to supply chain resilience, maritime safety, and environmental protection, given the sector’s critical role in global trade. The rapid adoption of MASS has outpaced existing governance frameworks, raising concerns about safety and regulatory legitimacy.

02openalex2026-07-31Y. Ding, W. Meng, S. He, Weiwei Li

Towards COLREGs-aware ship collision avoidance with multi-agent PPO-LSTM in maritime IoT (opens in a new tab)

This paper proposes a three-layer maritime IoT framework for autonomous ship collision avoidance, integrating vessel states, risk metrics, and environmental data to comply with COLREGs (International Regulations for Preventing Collisions at Sea). It uses a multi-agent PPO-LSTM algorithm with a centralised-training-decentralised-execution approach to handle partial observability and complex vessel interactions. Simulations in congested and narrow waterways show improved success rates, safer distances, and fewer rule violations compared to other methods. Explicit COLREGs-aware rewards and timely communication are key to its performance.

03openalex2026-07-29Cihad ÇelikDOI ↗ (opens in a new tab)

Incorporation of a System Identification–Based Ship Manoeuvring Model into Real-Time Collision Risk Analysis of Autonomous Ships (opens in a new tab)

This study introduces a data-driven framework to improve collision-risk assessment for autonomous ships by combining Bayesian risk modeling with system identification-based manoeuvring predictions. It analyzes global maritime accident data using Bayesian Networks to identify key risk factors, then applies Dynamic Bayesian Networks to predict real-time collision risks during ship encounters using AIS data. The research also develops a non-parametric model to predict ship manoeuvring behavior, validated with KCS benchmark ship data, achieving accuracy comparable to deep learning with lower computational costs. The findings support enhanced collision risk evaluation and autonomous navigation through realistic trajectory predictions and control strategies.

04openalex2026-07-28Simona Aracri, Fausto Ferreira, Angelo Odetti, Giorgio Bruzzone +14 moreDOI ↗ (opens in a new tab)

Citizens marine pilots: Increasing autonomous surface vehicles navigation capabilities with and for the public (opens in a new tab)

The paper discusses the MONUSEN project, an EU-funded initiative enhancing public engagement in marine robotics to support UN Sustainable Development Goal 14 (Life below Water). It involved two public activities: demonstrating MARUS, a simulator for realistic maritime environments, and allowing participants to pilot the SWAMP autonomous surface vehicle (ASV) for data collection. These activities successfully engaged diverse groups, improving public understanding of autonomous vehicle control and marine technology. The project bridged gaps between citizens and scientists, fostering interdisciplinary collaboration.

05openalex2026-08-01Yu Cai, Ying Li, L. Zhang, Jun SongDOI ↗ (opens in a new tab)

Dual-Agent Hierarchical Reinforcement Learning for Typhoon-Avoidance Route Planning of Ships Under Dynamic Wind–Wave–Current Fields (opens in a new tab)

This study introduces a dual-agent hierarchical reinforcement learning framework (HRL-MOO) to improve typhoon-avoidance ship routing under dynamic wind, wave, and current conditions. The model uses a global agent to assess large-scale environmental risks and a local agent to adjust short-term vessel movements, incorporating real-world bathymetric constraints. Simulations of Typhoon Yagi (2024) show the approach reduces environmental risk, improves path smoothness, and speeds up training convergence compared to single-agent methods. Ablation studies confirm the benefits of combining macro-level risk assessment with adaptive multi-objective optimization.

06openalex2026-08-02Md. Sabbir Ahmed, Zobair Ibn AwalDOI ↗ (opens in a new tab)

Autonomous Threat Assessment and Maneuver Control for Unarmed Ships Using ML-Based Pirate Detection (opens in a new tab)

This study presents a machine-learning maritime security system using YOLOv8 for pirate detection in low-light and adverse weather, trained on an augmented dataset. The system assesses threat levels by analyzing vessel size, speed, and heading to differentiate pirate skiffs from benign boats, reducing false positives. It calculates optimal evasive maneuvers, finding zigzag patterns most effective for unarmed ships. The framework operates in real-time on standard hardware, offering a proof-of-concept for early threat mitigation.

07openalex2026-07-29Daniel Autenrieth, Daniel Paland, Jan Mentjes, Adrian ZlockiDOI ↗ (opens in a new tab)

Toward Standardized Benchmarking of Search-Based Scenario Selection Methods in Autonomous System Validation (opens in a new tab)

This paper presents a benchmarking framework to standardize the evaluation of search-based scenario selection methods (SBSSMs) for validating autonomous systems. The framework offers a domain-independent environment with synthetic test instances, standardized metrics, and reference implementations to improve reproducibility and comparability. It abstracts application-specific details while preserving structural patterns from real-world scenarios, such as maritime simulations. The tool aims to facilitate method comparison and advance SBSSM development across fields like automotive safety.

08openalex2026-07-31Amit RayDOI ↗ (opens in a new tab)

Deployment, Manipulators, and Actuators (opens in a new tab)

The paper discusses the importance of Launch and Recovery Systems (LARS) for Unmanned Underwater Vehicles (UUVs), highlighting their role in mission operability and safety. It explores fixed, modular, and mobile LARS solutions, including autonomous docking systems for AUVs. The text also covers manipulator systems used in underwater intervention tasks, primarily by ROVs and AUVs. Emerging technologies like soft actuators and bio-inspired underwater robot designs are noted as areas of growing interest.

09openalex2026-07-29Mycelia Research TeamDOI ↗ (opens in a new tab)

Governance Geometry in Web3: The Vessel Compartment System as a Constitutional Ownership Framework (opens in a new tab)

The paper critiques token-weighted voting in Web3 for concentrating decision-making power and proposes the Vessel Compartment System (VCS) as an alternative. VCS divides governance into fixed, non-dilutable compartments for distinct stakeholder roles (e.g., community, operators, capital providers), maintaining balance even as participants change. Smart contracts enforce these boundaries, enabling stable yet adaptable governance through participant migration. The model aims to improve real-world economic coordination by embedding constitutional ownership structures in decentralized systems.

10openalex2026-08-01Rana Saha, Monica Lundh, Scott N. MacKinnonDOI ↗ (opens in a new tab)

Evaluating stakeholders’ interactions for future autonomous European inland waterway transport (opens in a new tab)

This study examines the shift toward autonomous systems in European Inland Waterway Transport (E-IWT) using a socio-technical framework. Researchers interviewed 29 stakeholders across four subsystems—Technical, Personnel, Work Design, and Environment—analyzing current operations and future scenarios, including manned and unmanned remote-control setups. Findings reveal a predicted shift from vessel-based to shore-based operations, requiring real-time communication, regulatory updates, and standardized Remote Operation Centers (ROCs). The study also highlights misalignments between subsystems and the need for stakeholder collaboration to support higher automation levels.

11openalex2026-07-29Xuehao Wang, L.R. Chen, Dongyang Xue, Cheng Chen +1 moreDOI ↗ (opens in a new tab)

Digital Twins for Marine Robotic Platforms: Architectures, Applications, and Challenges Across AUVs, ROVs, USVs, Underwater Gliders, and Emerging Platforms (opens in a new tab)

This paper reviews digital twin (DT) applications in marine robotics, covering AUVs, ROVs, USVs, underwater gliders, and emerging platforms. A structured search analyzed 107 studies, identifying varied architectures, technologies, and validation methods but no fully operational deployments. Key challenges include balancing model fidelity with computation, environmental uncertainties, and communication limitations. The authors suggest physics-informed learning and edge computing as potential solutions but highlight the need for sustained, interoperable physical-virtual integration.

12openalex2026-07-30Mengwei Chen, Guichen Zhang, Yuanlong Li, Bing HanDOI ↗ (opens in a new tab)

Finite-time composite dynamic compensation control for USVs with rotatable thrusters and multi-channel event-triggered inputs under thruster faults (opens in a new tab)

This paper addresses fault-tolerant control for underactuated unmanned surface vessels (USVs) with rotatable thrusters experiencing thruster faults. It introduces a dynamic compensation controller to manage input constraints and allocation errors, improving accuracy and convergence speed. A multi-channel event-triggered mechanism reduces unnecessary updates, saving network bandwidth, while a finite-time disturbance observer estimates faults and external disturbances. Theoretical analysis and simulations confirm the method’s stability and efficiency in fault compensation and communication.

13openalex2026-07-28Wenbo Luo, Qingwen Zhang, Bin LinDOI ↗ (opens in a new tab)

Research on the inspection technology and application of deep-sea marine cable based on underwater robots (opens in a new tab)

This paper examines the use of underwater robots to improve deep-sea marine cable inspection and maintenance. It evaluates ROVs, AUVs, and hybrid systems like UVMS, highlighting their roles in navigation, damage detection, and repairs. Advanced technologies such as vision-based systems, sonar, and multisensory fusion are discussed for enhancing accuracy in harsh conditions. Case studies demonstrate the effectiveness of these robotic platforms in real-world cable monitoring and repair tasks.

14openalex2026-07-29Tita Alissa Bach, Aleksandar Babic, Narae Park, Tor Sporsem +3 moreDOI ↗ (opens in a new tab)

Using LLM-generated draft replies to support human experts in responding to stakeholder inquiries in maritime industry: a real-world case study of Industrial AI (opens in a new tab)

This study explored the use of Large Language Models (LLMs) to help maritime industry experts draft responses to stakeholder inquiries. Researchers found LLMs improved efficiency, consistency, and time management but often required manual edits for accuracy. Experts showed cautious interest, stressing the need for human oversight and managerial support. While LLMs aren’t ready for independent use in safety-critical tasks, they can assist when combined with human expertise.

15openalex2026-07-29Xuefeng Qi, Wenfeng Liu, Fangyou Gong, Jianfeng Wu +3 moreDOI ↗ (opens in a new tab)

Transient Dynamic Analysis and Vibration Reduction Optimization of a Marine ROV Launch and Recovery System Based on Viscoelastic Damping (opens in a new tab)

This study analyzes the structural safety of a marine ROV launch and recovery system (LARS) under extreme sea conditions, focusing on its vulnerability to high-frequency impacts. Simulations revealed that traditional rigid frames face significant stress (195.38 MPa) and deformation (over 20 mm) under a 1 g horizontal impact. A viscoelastic damping solution was proposed, reducing peak stresses by over 37% and deformation by 64%. The approach offers a lightweight alternative to conventional strengthening methods for marine equipment.

© 2026 Ship Autonomy · Summaries are AI-generated; articles belong to their publishers · Privacy · Contact