(03)

Science

↓ (W15 ‘26 · Apr 6–12)
01openalex2026-04-09Joel Jose, Andreas Madsen, Andreas Brandsæter, Tor A. Johansen +1 moreDOI ↗ (opens in a new tab)

"Why This Avoidance Maneuver?" Contrastive Explanations in Human-Supervised Maritime Autonomous Navigation (opens in a new tab)

This paper examines how to improve transparency in maritime collision avoidance systems for human supervisors. It introduces contrastive explanations, comparing a system’s proposed maneuver with alternatives to clarify decision-making. A study with four marine officers found these explanations helpful, especially in complex multi-vessel scenarios, but noted increased cognitive workload. The authors suggest future interfaces should offer explanations on demand or tailored to specific situations.

02openalex2026-04-09Derar Al-DaboubiDOI ↗ (opens in a new tab)

The efficacy of the UAE legislation in regulating the liability of the maritime carrier in the era of automated shipping (opens in a new tab)

This paper analyzes the legal challenges of regulating liability for maritime carriers using autonomous vessels under UAE legislation. It focuses on defining the roles and responsibilities of the Shore-Based Controller (SBC), voyage programmers, and software providers, as well as addressing liability in cases of cyberattacks. The study recommends updates to international conventions and UAE Maritime Law (UAEML) to clarify liability for third-party actions and cyber threats. It also highlights gaps in UAEML regarding unseaworthiness caused by faults in software or hardware systems.

03crossref2026-04-11Xin Xiong, Rudy R. Negenborn, Yusong PangDOI ↗ (opens in a new tab)

Model predictive formation control of multi-vessel systems considering ship-to-ship interaction (opens in a new tab)

This paper introduces a centralized model predictive control (MPC) framework for autonomous surface vessels that accounts for ship-to-ship interactions, which are often ignored or oversimplified in traditional methods. The approach uses an empirically derived, three-degrees-of-freedom interaction model based on computational fluid dynamics to estimate proximity-induced forces in real time. Simulations show the MPC controller improves trajectory tracking and disturbance handling compared to PID controllers, particularly in platooning, parallel, and triangular formations. The study also examines performance variations across different formation types.

04crossref2026-04-08Elizabeth A. Suter, Christine Santora, Elizabeth T. Sciorilli, Natalia Benejam +10 moreDOI ↗ (opens in a new tab)

First Autonomous eDNA Survey Using a Self- Righting, Solar-Powered Surface Vehicle in an Estuarine Hope Spot (opens in a new tab)

This study deployed a solar-powered, self-righting surface vehicle equipped with an autonomous eDNA sampler in Shinnecock Bay, New York, a designated conservation area. Over eight days, the system collected 98 usable samples, revealing biodiversity patterns, including 15 priority conservation taxa and 8 invasive species, with higher occupancy in the less degraded eastern bay. Eelgrass beds were identified as key biodiversity hotspots, while nighttime sampling detected nocturnal species like anchovies and shortfin mako sharks. The approach supports scalable, emissions-free coastal monitoring aligned with global sustainability goals.

05openalex2026-04-08Kwangho Shin, Hyunwoo Song, Yonghoon Choi, Woongki Lee +2 moreDOI ↗ (opens in a new tab)

A Methodological Approach to Defining Seakeeping Performance Criteria for Unmanned Surface Vessel (opens in a new tab)

This paper proposes a method to define seakeeping performance criteria for unmanned surface vessels (USVs) to improve their operation in harsh maritime conditions. It reviews seakeeping data from various vessels, including manned ships, to develop a tailored procedure for USVs. A 7-meter-class USV serves as a case study, where preliminary criteria were established and validated through model tests. The goal is to enhance USV design for better performance in joint or independent naval missions.

06openalex2026-04-08S.B. Im, Si-Won Kim, Seonghyeon Jung, Ji Heui Seo +3 moreDOI ↗ (opens in a new tab)

A Study on RoI-based Data-efficient AtoN Classification for MASS using VLM (opens in a new tab)

This study introduces a data-efficient method for classifying Aids to Navigation (AtoN) using vision-language models (VLMs) and Region of Interest (RoI) processing for Maritime Autonomous Surface Ships (MASS). It addresses challenges like limited labeled data and background interference by comparing a YOLOv12 classifier with CLIP, which uses prompt engineering and few-shot tuning. Tests on simulated and real-world datasets show CLIP performs robustly with minimal training data, even in low-visibility conditions. The approach improves practical AtoN classification in maritime settings.

07crossref2026-04-07Shuang Liu, Lei Wei, Xiaoqing LiDOI ↗ (opens in a new tab)

CD-HSSRL: Cross-Domain Hierarchical Safe Switching Reinforcement Learning Framework for Autonomous Amphibious Robot Navigation (opens in a new tab)

The paper introduces CD-HSSRL, a reinforcement learning framework for autonomous amphibious robots navigating both water and land. It addresses challenges like discontinuous dynamics and safety risks by combining a global reachability planner, hierarchical switching policies, and a safety-constrained controller. Tests on multiple benchmarks show a 15% improvement in transition success and a 40% reduction in collisions compared to existing methods. The framework offers a unified solution for safe cross-domain navigation in amphibious robots.

08openalex2026-04-09Payal Mittal, Khagendra Saini, Anirudh Phophalia, Vaani MehtaDOI ↗ (opens in a new tab)

VAK-Former: Swin-L based Mask2Former for Maritime Semantic Segmentation (opens in a new tab)

VAK-Former is a semantic segmentation framework for autonomous Unmanned Surface Vessels (USVs) in maritime settings, using a Swin-L based Mask2Former architecture. It features a hierarchical vision transformer backbone, multi-scale pixel decoding, and query-based mask classification, optimized for the LaRS maritime dataset. The release includes reproducible training and evaluation scripts built on MMSegmentation, PyTorch 2.1.2, and Python 3.8. The code and associated manuscript are archived on Zenodo for long-term accessibility.

09openalex2026-04-06Zongrui Hao, Z. Ping, K. Li, J. Xu +3 moreDOI ↗ (opens in a new tab)

An Investigation on Wake Control Mechanism and Performance Optimization of the Flow-interrupting Plate in Unmanned Sailing Vessels (opens in a new tab)

This study examines the use of stern-mounted flow-interrupting plates to reduce wake disturbances in unmanned sailing vessels, which impact efficiency and stealth. Using CFD simulations with the SST k–ω turbulence model and VOF method, researchers tested varying plate angles and positions. A 2.5° inclination and 10 mm waterline distance minimized vortex diffusion, cut wake height, and limited added resistance to ~4 N. The plate achieved wave height reductions of up to 57.08% in downstream regions, offering a practical solution for wake suppression and stealth optimization.

10openalex2026-04-10Ho-Chih Cheng, Min-Chie Chiu, Ming-Guo HerDOI ↗ (opens in a new tab)

Optimization of a Ship-Based Three-Magnet Energy Harvester Using Wave Excitation via the Flower Pollination and Simulated Annealing Algorithms (opens in a new tab)

This study explores a ship-based energy harvester using wave-induced vibrations to power offshore underwater sensors. A three-magnet system, mounted above sea level for safety and simplicity, converts wave motion into electrical energy. Optimization via the flower pollination and simulated annealing algorithms improved power output, achieving up to 0.1943 W under tested wave conditions (3.0 m/s speed, 0.4 m amplitude, 2.0 m wavelength). The system’s design avoids underwater risks and shows promise as a sustainable power source for marine equipment.

11openalex2026-04-08Xuanchen LiDOI ↗ (opens in a new tab)

"Lidar_Ship_detectedBY_MID360" (opens in a new tab)

This paper presents a 3D point cloud dataset collected using a Livox Mid-360 LiDAR sensor, targeting real-world vessel detection on inland lakes. It includes small fishing boats with detailed hull structures, water reflections, and varied environmental conditions, formatted for direct use with the OpenPCDet framework. The dataset contains raw point clouds, annotated 3D bounding boxes, and pre-processed metadata for training and validation. It supports research in 3D object detection, segmentation, and maritime perception for unmanned surface vehicles.

12crossref2026-04-07Jose Vicente Roig, Julian SaltDOI ↗ (opens in a new tab)

Autonomous Vehicle Front Steering Control Computation Saving (opens in a new tab)

This paper explores a method to reduce computational demands in autonomous vehicle steering control. Traditional robust controllers for lane-keeping systems often result in high-order computations, straining processing resources. The authors propose an interlacing technique to simplify the controller’s implementation, lowering computational load while maintaining performance. The approach is tested on real path-tracking scenarios, demonstrating its effectiveness for state-space and MIMO controllers.

13openalex2026-04-09Martín Fuertes OlivaDOI ↗ (opens in a new tab)

Black Cruiser: A Fractal-Palindromic Vessel for Autonomous Energy Navigation (OPEN SOURCE - FREE PATENTS -OLD MISTAKED DRAFT) (opens in a new tab)

The *Black Cruiser* is a theoretical spacecraft concept combining the *Black Motor* and the electromagnetic hypothesis of black as a palindromic fractal. Its energy core uses symbiotic transduction to convert ambient and vacuum radiation into thrust and control energy. Designed for long-range navigation, it features modular scalability, fractal curvature manipulation, and multidimensional folding. The draft is open-source and patent-free.

14crossref2026-04-06S. Krithikka, G. Hariharan, H. JafariDOI ↗ (opens in a new tab)

Fermat polynomial-based machine learning algorithm for a few non-linear ship roll damping models arising in ship dynamics (opens in a new tab)

This study introduces a machine-learning model using Fermat polynomial methods (FPM) to predict ship roll angles by solving non-linear differential equations in ship dynamics. The approach transforms complex equations into algebraic forms and incorporates restoring moments and damping coefficients for FPSO and barge-like vessel models. Accuracy was validated against experimental data and the homotopy perturbation method (HPM), showing FPM as a simpler and effective alternative. Results were compared with numerical methods and real-world observations.

17crossref2026-04-07İsa Ünver, Musa DemirciDOI ↗ (opens in a new tab)

Design Of an Unmanned Surface Vehicle and Flow and Structural Analysis of Its Hull (opens in a new tab)

This study presents the design and analysis of a 7-meter V-type trimaran hull for an Unmanned Surface Vehicle (USV) using SolidWorks and ANSYS. The research combines computational fluid dynamics (CFD) and structural analysis to assess hydrodynamic efficiency, stability, and resistance to external forces, including stress and deformation. Modal analysis was also conducted to evaluate the hull’s response to environmental vibrations. The integrated approach aims to optimize USV performance for high-speed and rough sea conditions while supporting defense industry applications.

18crossref2026-06-01Rohan Chandra, Shubham Singh, Wenhao Luo, Katia SycaraDOI ↗ (opens in a new tab)

Multi-robot navigation in social mini-games: definitions, taxonomy, and algorithms (opens in a new tab)

This paper introduces "Social Mini-Games" (SMGs) as constrained, high-agency environments (e.g., doorways, hallways) where robots must navigate while interacting with humans and other robots. It highlights inconsistencies in assumptions, objectives, and evaluation methods across existing SMG navigation research, creating challenges for comparisons and practical applications. The authors propose a unified taxonomy and definitions to standardize the field, classify current methods, and outline evaluation protocols. The work aims to guide future research and lower barriers for new researchers in multi-robot navigation.

19openalex2026-04-07Mrs.KanakaTulasi P.ReddiDOI ↗ (opens in a new tab)

SeaGuard-AI: An Adversarial Robust Framework For Reliable Sea State Estimation In Autonomous Marine Vessels (opens in a new tab)

The paper *SeaGuard-AI* introduces an adversarial-robust framework designed for reliable sea state estimation in autonomous marine vessels. Authored by Mrs. Kanaka Tulasi P. Reddi, it focuses on improving the accuracy and security of image-processing-based systems against adversarial attacks. The work is scheduled for publication on April 7, 2026. Additional unrelated topics, such as VLSI surveys and free paper downloads, are mentioned but not part of the core research.

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