Ship Autonomy

01openalex2026-10-28Sergey Victorovich Smolentsev, А. А. Мироненко, A.E. FilyakovDOI ↗

A comprehensive model for assessing the risk of close approach between maritime autonomous surface ships and objects (opens in a new tab)

This study enhances the ship domain model using an N-focal ellipse to improve risk assessment for maritime autonomous surface ships (MASS). It distinguishes between collision-avoidance tasks involving moving objects under COLREGs and other navigational hazards by categorizing foci into vessel dimensions, kinematic parameters, and maneuvering characteristics. The model introduces multipliers to adjust domain boundaries based on navigation conditions and defines a minimum navigational safety margin. Simulations support the findings, suggesting the model could advance MASS navigational safety when integrated into control systems.

02openalex2026-07-15Sojin Kim, Jeonghong ParkDOI ↗

BIT* -based Path Planning for Remotely Operated Maritime Autonomous Surface Ships Under Dynamic Obstacle Uncertainty (opens in a new tab)

This paper introduces a BIT*-based path-planning method for Maritime Autonomous Surface Ships (MASS) to handle dynamic obstacle uncertainties in congested waters. The approach combines collision avoidance and path replanning into a single optimization problem, using probabilistic collision risk in the BIT* cost function to prioritize safer routes. It also integrates COLREGs rules to ensure compliance with maritime navigation regulations. Simulations show the method produces safe and reliable paths despite positional uncertainty from sensor errors and communication delays.

03openalex2026-10-28Victor Vladimirovich Deryabin, Сазонов Анатолий ЕфимовичDOI ↗

A neural network-based approach to collision avoidance for maritime autonomous surface ships (opens in a new tab)

This paper proposes a neural network-based method for cooperative collision avoidance in Maritime Autonomous Surface Ships (MASS). The system uses vessel coordinates and velocity components as inputs and outputs course adjustments to maintain a safe distance. The neural network has eight hidden layers with 1,000 neurons each and was trained on one million samples using a two-iteration gradient optimization process. After training, the model achieved a mean prediction error of about 2°, though maximum errors remained high at several tens of degrees.

04openalex2026-09-24Yucheng Zhang, Dezhi Wang, Xiaori Gao, Jiayi Che +2 moreDOI ↗

A Decoupled Global-Local Planning Framework for COLREGs-Aware Maritime Autonomous Surface Ships in Coastal Waters (opens in a new tab)

This paper introduces a decoupled framework for COLREGs-compliant autonomous ship navigation in coastal waters, separating global path planning from local collision avoidance. The global layer uses Adaptive Angular-Domain Potential Field (AAPF) for path generation and Farsighted Sweep Trajectory Optimization (FSTO) for refinement, while the local layer employs Environment-Adaptive Dynamic Window Approach (EA-DWA) for real-time risk assessment and avoidance. Simulations demonstrate reduced computational burden and improved path usability, though performance depends on environment-specific parameter adjustments. The approach ensures compliance with maritime collision regulations while balancing efficiency and adaptability.

05crossref2026-08-31Ke Zhang, Jie Wen, Xingya Zhao, Jianan Luo +1 moreDOI ↗

A ship autonomous navigation decision-making method based on the OODA loop theory (opens in a new tab)

This study introduces an autonomous ship navigation method using the OODA (Observe–Orient–Decide–Act) loop theory to improve safety and decision-making in complex maritime scenarios. The approach includes dynamic ship domain modeling, risk assessment, and collision avoidance strategies based on a three-degree-of-freedom motion model and an enhanced velocity obstacle algorithm. Simulations on OpenCPN confirmed the method’s effectiveness, maintaining safe distances and compliance with COLREGs (Collision Regulations) even in high-risk encounters. The framework provides a closed-loop system for real-world autonomous navigation applications.

06crossref2026-07-07Hiba Ben Lahbib, Mohamed Taha Bennani, Jüri VainDOI ↗

Autonomous vessel collision avoidance verification using geometric optimization (opens in a new tab)

This paper addresses gaps in the International Regulations for Preventing Collisions at Sea (COLREG) for Maritime Autonomous Surface Ships (MASS) by incorporating relative speed variations between vessels, expanding on prior work limited to identical-speed scenarios. It adapts collision-risk geometry calculations, such as collision angle, to assess how speed differences affect navigation safety. The study refines a geometric optimization method to efficiently analyze two-dimensional encounter scenarios, maintaining near-polynomial search complexity. The approach identifies high-risk vessel positions in under half a second, making it viable for onboard navigation warning systems.

07crossref2026-04-29Natanael Suranta, Tri Cahyadi, Irfan FaozunDOI ↗

Legal Accountability of Deck Officers in Maritime Autonomous Surface Ships: A Normative Analysis of COLREGS and STCW Compliance Gaps in Indonesian Domestic Shipping (opens in a new tab)

This study analyzes legal gaps in regulating Maritime Autonomous Surface Ships (MASS) under existing frameworks like COLREGS, STCW, and Indonesian Law No. 17 of 2008. It identifies three key issues: undefined legal responsibility for autonomous decisions, outdated STCW standards for remote operations, and unclear liability among flag states, operators, and tech developers. Using legal and comparative analysis, the research proposes a layered accountability model for Indonesia’s domestic shipping sector. The findings aim to guide regulatory reforms for MASS integration.

08openalex2026-09-18Lu Liu, Yuan He, Xinwei Liu, Zhijian Sun +1 moreDOI ↗

Human-in-the-loop COLREGs-compliant platoon formation control of maritime autonomous surface ships via finite state machine (opens in a new tab)

This study presents a control system for maritime autonomous surface ships (MASSs) that integrates human supervision into platoon navigation while adhering to COLREGs collision avoidance rules. A virtual leader allows human operators to adjust the platoon’s speed, while a finite state machine identifies encounter scenarios and applies safety constraints via control barrier functions. The system uses a cooperative path-following guidance law and quadratic optimization to balance collision avoidance with platoon coordination. Simulations confirm the approach effectively maintains formation and complies with maritime safety regulations.

09openalex2026-08-19P. Maes, Erin R. McMahon, Doreen Jirak, Dirk van Rooy +4 moreDOI ↗

Towards Human Factors Evaluation of Interactions with Maritime Autonomous Surface Ships Using Full Mission Bridge Simulators: The Conventional Officer’s Perspective (opens in a new tab)

This study examines how Maritime Autonomous Surface Ships (MASS) interact with crewed vessels, focusing on the challenges faced by conventional Officers of the Watch (OOW) during collision avoidance. It introduces a framework analyzing situational awareness, mental models, and decision-making in mixed-traffic scenarios. Key issues include communication gaps, uncertainty about MASS capabilities, and trust in autonomous systems. The paper proposes using Full Mission Bridge Simulators to study OOW behavior and decision-making in these encounters.

10crossref2026-07-08Linna Li, Lingyu Zhang, Seyed Parsa Parvasi, Jasmine Siu Lee LamDOI ↗

Machine learning-based maritime collision risk assessment and avoidance for autonomous vessels using AIS data (opens in a new tab)

This study proposes a machine learning framework using XGBoost to assess maritime collision risks for autonomous vessels, replacing the computationally intensive Dempster-Shafer model. Using AIS data, the model predicts Collision Risk Indicators (CRI) with high accuracy (95.36% R² in training, 90.21% in testing) and integrates them into collision avoidance algorithms. The approach processes high-traffic scenarios faster than traditional methods and reduces collision risk by 33.0% in autonomous-autonomous encounters and 28.7% in autonomous-conventional encounters. Results highlight improved efficiency and scalability for real-time maritime navigation safety.

11openalex2026-03-30S. Sun, Liangbin Zhao, Ming Deng, Xiuju FuDOI ↗

From Vessel Trajectories to Safety-Critical Encounter Scenarios: A Generative AI Framework for Autonomous Ship Digital Testing (opens in a new tab)

The paper introduces a generative AI framework that converts real-world ship trajectories from AIS data into realistic safety-critical encounter scenarios for testing autonomous maritime navigation systems. It uses a multi-scale temporal variational autoencoder to improve trajectory realism and robustness, even with noisy AIS data. The method enables scalable generation of diverse high-risk scenarios beyond historical records, supporting digital testing and safety assessments. Code for the framework is publicly available.

12crossref2026-10-22Liang He, Xiong Xiao, Mingjie Li, Weijun WangDOI ↗

PSN-PPO: A Policy-Splitting Network-Based Deep Reinforcement Learning Approach for Autonomous Navigation of Unmanned Surface Vehicles (opens in a new tab)

The paper introduces PSN-PPO, a deep reinforcement learning method for autonomous navigation of unmanned surface vehicles (USVs) that separates navigation and obstacle avoidance into two jointly trained policy branches. The approach uses a navigation branch for course guidance and an avoidance branch for bounded corrections, modulated by a learnable risk gate. Simulations across five scenarios—including obstacle-free navigation, static obstacles, and dynamic encounters—show PSN-PPO improves efficiency and collision avoidance compared to standard PPO. The method was tested using a three-degree-of-freedom vessel motion model in a closed-loop framework.

13openalex2026-07-31Y. 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.

14crossref2026-07-07Xiaoru Ma, Hongguang Lyu, G. H. Tan, Xikai Kang +2 moreDOI ↗

Autonomous collision avoidance for complex multi-ship encounters based on improved dynamic window approach (opens in a new tab)

This study enhances autonomous ship navigation by improving collision avoidance in complex multi-ship encounters. It combines Velocity Obstacle (VO) and CPA/TCPA methods to assess collision risks and applies COLREGs rules to assign give-way and stand-on responsibilities. The proposed MVO-IDWA algorithm integrates these factors into the Dynamic Window Approach to optimize decision-making. Results show the method effectively considers ship responsibilities, risk levels, and movement trends for safe navigation.

15openalex2026-06-20Vishnu RavendranathanDOI ↗

VesselNav-Bench: A Transparent, Reproducible Benchmark for Comparing Autonomous Ship Navigation Policies (opens in a new tab)

VesselNav-Bench is an open benchmark for comparing autonomous ship navigation policies, using a shared 3-DOF simulator to ensure transparent and reproducible evaluations. Seven agents, including rule-based, optimization-based, and learning-based methods, were tested across nine scenarios under calm and disturbed conditions, with 360 episodes per agent. Classical and MPC approaches performed best in calm conditions, while MPC excelled under disturbances; safety shielding improved RL policy safety but revealed persistent generalization gaps. COLREGS Rule 17 compliance was limited to 0.76 across all agents, highlighting challenges for ASV certification. The benchmark, including code and logs, is publicly available.

16openalex2026-06-08Hosna Namazi, Lokukaluge P. PereraDOI ↗

Learning human collision avoidance behavior for autonomous ships using adaptive neuro-fuzzy inference system (opens in a new tab)

This study introduces an Adaptive Neuro-Fuzzy Inference System (ANFIS) to improve collision avoidance for autonomous ships by mimicking human navigator behavior. The system uses fuzzy clustering (FCM) and high-fidelity simulator data to train separate models for crossing, head-on, and overtaking scenarios, ensuring compliance with COLREGs and safe maneuvering. Validation shows smooth, collision-free trajectories with rudder commands resembling human actions. Sensitivity analysis reveals that different encounter types rely on distinct input variables, such as geometric factors for crossing and overtaking.

17openalex2026-09-28Aqsa Yaseen, Marvin Roshan Crompton Ramesh, Karl Janson, Pentti KujalaDOI ↗

A Review: Learning-Based Collision Avoidance Methods for Autonomous Vessels (2021–2025) (opens in a new tab)

This review examines machine learning methods for autonomous vessel collision avoidance from 2021 to 2025, analyzing 445 studies after screening 832 records. Reinforcement learning, particularly Deep Reinforcement Learning (DRL), Proximal Policy Optimization (PPO), and Deep Q-Network (DQN), emerged as the leading approaches. The adoption of these techniques surged after 2022, reflecting their growing role in adaptive decision-making and trajectory prediction. The paper also outlines challenges in real-world deployment and future research directions.

18openalex2026-07-29Cihad ÇelikDOI ↗

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.

19openalex2026-07-24Ke Zhang, Xiongfei Geng, Jie Wen, Liu X +1 moreDOI ↗

Adaptive collision-avoidance timing recognition for autonomous ship encounters using multi-evidence Bayesian backward evidence detection (opens in a new tab)

This study introduces a Bayesian framework to improve collision-avoidance timing for autonomous ships by analyzing historical AIS data. It integrates maneuvering, motion, and risk evolution to detect adaptive trigger thresholds for different encounter types (overtaking, head-on, crossing). Tests on 6,186 AIS encounters showed avoidance actions began 6.5–8.6 minutes before peak collision risk, with adaptive thresholds ranging from 0.245 to 0.368. The method outperformed fixed-threshold approaches in stability and interpretability.

20crossref2026-05-07L.P. PereraDOI ↗

Sandbox environment for evaluating safety-critical future autonomous ship navigation (opens in a new tab)

This paper examines the limitations of current autopilot systems in autonomous ships, which mainly handle basic navigation and lack strong collision avoidance. It proposes using VR simulators and model-scale vessel testing as a controlled "sandbox" environment to improve autonomous navigation. The study highlights the role of advanced predictors and AI in detecting and preventing close-encounter situations early. It also stresses the need to ensure both hardware and software reliability in AI systems to handle potential failures and maintain safety.

21openalex2026-07-01Manlin Wang, Hongjun Tian, Maoyuan Sun, Yuhan Zhou +10 moreDOI ↗

Bridging Symmetric Dynamics and Asymmetric Semantic Objectives: Runtime-Assured Predictive Safety Control for Autonomous Surface Vehicles (opens in a new tab)

This paper introduces a predictive safety-control framework for autonomous surface vehicles (ASVs) that reconciles symmetric vessel dynamics with asymmetric collision-avoidance rules like COLREGs. A large language model (LLM) generates and validates structured COLREGs labels to guide asymmetric responsibilities, while a physical safety envelope, built on motion prediction and uncertainty inflation, ensures collision avoidance. A runtime assurance monitor checks semantic validity and solver performance to prevent unsafe decisions. Simulations show the approach improves rule-compliant navigation while reducing overly cautious maneuvers.

22openalex2026-10-11Minjie Zheng, Yuanhao Zhou, Zunyuan Wang, Guoquan Chen +1 moreDOI ↗

A Hierarchical Logic-Guided Reinforcement Learning Framework for Interpretable Autonomous Ship Collision Avoidance (opens in a new tab)

This paper introduces the Logic-Guided Interpretable Deep Reinforcement Learning (LG-IDRL) framework for autonomous ship collision avoidance, combining COLREGs-based rule reasoning with reinforcement learning (PPO). The system dynamically adjusts rewards and action constraints to ensure compliance with maritime rules while optimizing navigation. In tests, LG-IDRL outperformed a baseline, achieving a 92.9% success rate, 1.6% collision rate, and 98.7% rule compliance. Results are limited to predefined two-ship encounters and do not guarantee broader applicability.

23openalex2026-09-11Dominik Frey, Ulf Kargén, Dániel VarróDOI ↗

Automated Generation of Functionally Complete Assurance Suites for COLREGS-Compliance of Autonomous Surface Vehicles (opens in a new tab)

This paper presents a model-based method to automatically generate assurance scenarios for verifying COLREGS compliance in autonomous surface vehicles (ASVs). It derives functional equivalence classes for multi-vessel encounters, maps them to logical scenarios with geometric constraints, and produces initial scenes using search-based and rejection sampling algorithms. The approach ensures full coverage of COLREGS-relevant encounters and scales to scenarios involving up to six vessels. Evaluation confirms its effectiveness for ASV assurance in open-sea conditions.

24openalex2026-10-13Won-Jun Yoo, Kwang-Jun Paik, Mu-Yeong SeoDOI ↗

Real-time situational awareness and collision risk evaluation for autonomous ships under sensor uncertainty (opens in a new tab)

This study introduces a framework for autonomous ships to assess collision risks in real time, accounting for sensor uncertainty from AIS and radar data. It uses an Unscented Kalman Filter to estimate target-ship states and integrates uncertainty into a probabilistic ship domain, enabling metrics like Probabilistic Collision Exposure (PCE) and Expected Collision Energy (ECE). Simulations under noisy conditions showed target-position errors averaging 3 m, with PCE reducing post-CPA residual risk significantly compared to the Collision Risk Index. The framework maintained a 10-Hz update rate across single- and multi-target scenarios.

25Splash2472026-11-02

US investigators call for autonomous vessel rules after Lake Erie collision (opens in a new tab)

The NTSB recommended new U.S. safety standards for uncrewed vessels after a 2024 collision in Lake Erie between a NOAA autonomous research craft and a recreational yacht. The low visibility and passive remote monitoring of the uncrewed vessel contributed to the incident, causing significant damage and minor injuries. Investigators found no existing industry-wide regulations for detectability, collision avoidance, or navigational safety of such vessels. The NTSB urged the U.S. Coast Guard to create a national regulatory framework to address these gaps.

26crossref2026-07-10Karim HardyDOI ↗

Engineering Safety Requirements for Maritime Autonomous Surface Systems: Hazard Scenarios, Control Loss, and Recovery in Degraded Operations (opens in a new tab)

This paper examines safety challenges in maritime autonomous surface systems, highlighting risks from degraded communication, unclear control authority, and delayed human intervention. It analyzes 80 documented hazard scenarios from regulations, trials, and incident reports to identify key risk areas like remote supervision, control links, and navigation interactions. The study proposes a safety engineering approach to define control constraints and recovery requirements, emphasizing the need for explicit degraded modes, fallback behaviors, and emergency coordination. Findings show safety depends more on system design and recovery protocols than autonomy level alone.

27crossref2026-06-30Xiaohui Wang, Yingjun Zhang, Shaobo Wang, Zhiyuan JiangDOI ↗

A negotiation method for cooperative collision avoidance between ships in mixed navigation scenarios: model construction, strategy driving, and unilateral learning (opens in a new tab)

The study introduces a bilateral negotiation model to enable autonomous and conventional ships to coordinate collision avoidance in mixed navigation scenarios. Using the Zeuthen strategy for bargaining and unilateral Bayesian learning for information estimation, the method improves negotiation efficiency with minimal data exchange from conventional ships. Simulation results show it resolves uncoordinated encounters, reduces unnecessary evasive maneuvers by autonomous ships, and enhances navigational safety. The approach addresses gaps in collaborative collision avoidance for heterogeneous fleets.

28openalex2026-10-21Stephen Monnet, Joanna Szłapczyńska, Simon Blindheim, Tor Arne Johansen +1 moreDOI ↗

Ship collision avoidance using timespace parametrization: Robust and risk-aware planning Using SeaCharts implementation (opens in a new tab)

This paper presents a COLREGs-compliant trajectory planning framework for autonomous ships that unifies static and dynamic obstacle avoidance by parameterizing time in space. It converts dynamic obstacles into polygonal representations, integrating global and local planners while accounting for uncertainties in vessel speed and course. The method is implemented using an extended version of the SeaCharts Python library, which now supports S-57 files and weather data. Case studies demonstrate improved robustness against target ship motion uncertainty compared to velocity obstacle and scenario-based MPC approaches.

29openalex2026-10-05Min Zhang, 郭沐壮, Yiyang ZouDOI ↗

Identification of key risk ships for MASS collision avoidance in mixed traffic through axiomatic fuzzy set-based semantic clustering (opens in a new tab)

This study introduces a method to identify high-risk ships in mixed maritime traffic involving autonomous (MASS) and conventional vessels. It uses a 4D-OPTICS approach to filter ships in dense encounters and an Axiomatic Fuzzy Set (AFS) model to convert risk indicators into interpretable semantic categories like spatial proximity and motion uncertainty. A hierarchical model then prioritizes ships based on COLREGs rules and semantic risk levels. Simulations and real-world AIS data from Yingkou Port validate the framework’s effectiveness.

30openalex2026-09-18Anastasiya GolovinaDOI ↗

Compliance with International Regulations for Preventing Collisions at Sea in the Context of the Introduction of Autonomous Shipping (opens in a new tab)

The study analyzes compliance with international collision prevention regulations, focusing on lights and signs, as autonomous shipping emerges. It highlights gaps in applying existing rules—designed for human-operated vessels—to autonomous ships, which rely on automated systems. The paper argues that autonomous vessels must meet equivalent safety standards and proposes legal and technical improvements to address enforcement challenges. Recommendations include adopting new technologies to ensure proper signaling on autonomous ships.

31openalex2026-08-28Andrés Yair Carvajal Bolivar, Fernando Jesus Guevara CarazasDOI ↗

Real-Time Obstacle Avoidance System for Autonomous Vessels Using Sensor Fusion (opens in a new tab)

This paper details a real-time obstacle avoidance system for autonomous vessels, tested in a 2D maritime simulation. The system uses a proximity detection algorithm built with Pygame to adjust ship trajectories and prevent collisions with static or moving obstacles. It incorporates a ship motion model and particle system for realistic environmental representation. The approach aims to support autonomous navigation, telemetry control, and intelligent maritime transport solutions.

32openalex2026-08-12Yuqing Lin, Rangya Zhang, Kum Fai YuenDOI ↗

IoT-Enabled Autonomous Maritime Navigation in Smart Ports: A Curriculum-Guided Shared Policy Learning Framework (opens in a new tab)

This paper presents a curriculum-guided reinforcement learning framework for IoT-enabled autonomous maritime navigation in smart ports. The approach uses a shared recurrent policy to improve decision-making under partial observability and dense traffic, with centralized offline training but fully onboard execution. Simulations in realistic port environments show improved reliability, collision avoidance, and training stability over baseline methods. The method also generalizes well to unseen high-density scenarios, offering a scalable solution for smart port operations.

33crossref2026-07-17Murat Selçuk SolmazDOI ↗

Collision Liability for Autonomous Vessels (opens in a new tab)

The paper examines legal challenges arising from collisions involving autonomous vessels, or Maritime Autonomous Surface Ships (MASS), as their use grows alongside manned ships. The International Maritime Organization (IMO) has defined MASS and categorized their autonomy levels, but liability rules for collisions remain unclear. The study analyzes potential liability issues and proposes recommendations to establish a legal framework for such incidents. Current regulations only address collisions between manned vessels, leaving gaps for autonomous vessel scenarios.

34openalex2026-06-25Won-Jun Yoo, Kwang-Jun Paik, Mu-Yeong SeoDOI ↗

Real-time Collision Avoidance Considering Ship Position Uncertainty Based on AIS Data (opens in a new tab)

This study presents a collision avoidance framework for autonomous ships that accounts for positional uncertainty in AIS data. An unscented Kalman filter estimates the motion and uncertainty of nearby vessels in real time, improving trajectory predictions. Collision risk is assessed using a quantitative index, and avoidance maneuvers are generated via the velocity obstacle method. Simulations demonstrate a 96% reduction in estimation error, achieving ~3 m positional accuracy compared to nonpredictive approaches.

35crossref2026-05-29Lirong Kou, Xiaoyang GaoDOI ↗

Advances in Optimized and Safe Path Planning of Marine Autonomous Surface Vehicles: A Review (opens in a new tab)

This review examines recent advancements in path planning for Autonomous Surface Vehicles (ASVs), highlighting five key issues: navigation environment, environment modeling, ship motion models, collision avoidance, and optimization. It categorizes existing algorithms into graph search-based, sampling-based, numerical optimization-based, and AI-based methods, detailing their improvements and applications. Traditional algorithms excel in static environments but struggle with dynamic constraints, while AI-based approaches, particularly deep reinforcement learning, show superior adaptability and multi-agent coordination. The paper identifies current challenges and outlines future research directions for safer and more optimized ASV navigation.

36openalex2026-10-26Reyes Poo Argüelles, Jesús A. García MazaDOI ↗

Algorithms for determining deadlines and COLREG-compliant manoeuvres for safe navigation in multi-ship scenarios using velocity areas (opens in a new tab)

This paper introduces algorithms for automatic collision avoidance in multi-ship scenarios by identifying risk-free navigation areas and compliant maneuvers. Using AIS data, safety parameters, and COLREG rules, the algorithms replicate human decision-making processes of navigational officers. Polar graphs and weighted velocity areas help determine optimal maneuvers while adhering to maritime regulations. The output is designed for integration into autonomous navigation systems.

37crossref2026-09-11Xiaojie ZhangDOI ↗

Safety Analysis of Collision Avoidance for Semi-Autonomous Navigation Ship Based on STPA-CREAM (opens in a new tab)

This study analyzes safety risks in semi-autonomous navigation ships using System-Theoretic Process Analysis (STPA) to model interactions in encounter scenarios and identify potential hazards. To strengthen human factors analysis, the Cognitive Reliability and Error Analysis Method (CREAM) is integrated with STPA. The combined STPA-CREAM approach provides targeted recommendations to enhance the safety of semi-autonomous navigation systems. The research focuses on improving navigation efficiency while addressing risks in intelligent ship operations.

38openalex2026-04-09Joel Jose, Andreas Madsen, Andreas Brandsæter, Tor A. Johansen +1 moreDOI ↗

"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.

39openalex2026-10-29Qinghua Zhu, Shenping Hu, JiaXin Du, Jing Xin Xu +1 moreDOI ↗

Intelligent anti-collision in multi-vessel scenarios: a spatial DPAD decision-making model via accident data-driven (opens in a new tab)

This paper introduces a **Domain Predicted Area of Danger (DPAD)** model for intelligent anti-collision decision-making in multi-vessel scenarios. The model constructs a **Predicted Collision Risk Area (PCRA)** based on COLREGs guidelines and uses **Potential Collision Risk (PCR)** thresholds to evaluate safe navigation resumption. **Grey Wolf Optimization (GWO)** and **Bézier curves** optimize post-collision recovery paths. Validated with real collision cases, the model demonstrates reliable adaptive control for vessel avoidance in complex waters.

40openalex2026-09-30Huu Quyen NguyenDOI ↗

TRAJECTORY TRACKING WITH COLLISION AVOIDANCE FOR AUTONOMOUS SURFACE VESSELS IN DYNAMIC OBSTACLE ENVIRONMENTS USING TRAJECTORY PREDICTION AND ADAPTIVE SLIDING MODE CONTROL (opens in a new tab)

This paper introduces a control system for autonomous surface vessels that handles both trajectory tracking and collision avoidance in dynamic environments. It uses Time to Closest Point of Approach (TCPA), Distance at Closest Point of Approach (DCPA), and Collision Risk Index (CRI) to assess risks and adjust the vessel’s path in real time. An Adaptive Sliding Mode Control (ASMC) ensures precise tracking and resilience against disturbances without needing prior knowledge of their limits. Simulations confirm the method’s effectiveness in avoiding collisions while maintaining stability and accuracy.

41openalex2026-09-21Ahui Xu, Yuanqiao Wen, Wenjun Tang, Yamin Huang +3 moreDOI ↗

MBSE-based development and verification of function-oriented ODDs for autonomous ships (opens in a new tab)

This study introduces an MBSE-based method to define, quantify, and verify operational design domains (ODDs) for autonomous ship functions. It converts regulatory, mission, and risk requirements into measurable ODD conditions, ensuring traceability across system architecture, verification scenarios, and runtime logic. The approach is demonstrated through a collision-avoidance case study, where ODD conditions are formalized and tested via model checking, simulations, and risk analysis. Results confirm consistency in model-level verification but do not validate real-world operational safety or threshold effectiveness.

42crossref2026-06-30Jae-Yong Lee, Ho Namgung, Joo-Sung KimDOI ↗

Development of collision-case-based testing scenarios for validating autonomous ship collision-avoidance algorithms (opens in a new tab)

This study introduces a framework for creating realistic collision-avoidance testing scenarios for autonomous ships by analyzing actual collision cases. The framework involves collecting collision data, extracting ship trajectories, and developing diverse encounter situations by adjusting ship roles and positions. The resulting scenarios reflect real-world complexities, such as rule violations, dynamic encounters, speed changes, and environmental factors. These scenarios aim to improve the validation of collision-avoidance algorithms and enhance maritime safety.

43openalex2026-05-31L. Vagushchenko, A. Kozachenko, J. KozachenkoDOI ↗

Interpretation and updating of regulations for preventing collision at sea (opens in a new tab)

This paper examines updates needed for the International Regulations for Preventing Collisions at Sea (COLREGS) due to advancements like larger and faster ships, new navigation technologies, and autonomous vessels. It proposes methods to interpret COLREGS, including defining danger zones and determining timely actions for ships. The authors critique existing rules, suggest amendments, and emphasize classifying vessel approach situations. The study also addresses incorporating regulations for Maritime Autonomous Surface Ships (MASS) into COLREGS.

44portnews.ru2026-10-22

South Korean province seeks autonomous ship zone off Geoje for Samsung Heavy trials (opens in a new tab)

South Korea’s South Gyeongsang Province has requested approval to designate waters off Geoje Island as an autonomous-ship operating zone, enabling Samsung Heavy Industries to continue sea trials after its current regulatory exemption ends in 2026. The proposed area covers coastal waters near Hansan, Gajo, and Chilcheon islands, with a decision expected by year-end. Samsung Heavy is testing a 12-meter autonomous vessel featuring AI-based navigation, collision avoidance, and sensor systems. The province aims to support the technology while minimizing disruptions to fishing, shipping, and leisure activities.

45openalex2026-10-21Fengde Qu, Yiting Gong, Jingjing Liu, Hongguang Lyu +1 moreDOI ↗

A Review of Ship Path Planning: Models, Algorithms, and Challenges (opens in a new tab)

This paper reviews ship path planning (SPP) advancements, highlighting its role in improving navigation safety and energy efficiency for autonomous ships. It categorizes modeling frameworks, algorithms, and evaluation metrics while addressing challenges like COLREGs compliance, path smoothing, and uncertainty. Hybrid algorithms combining global optimization with real-time local adjustments are identified as the leading solution. Future progress depends on deeper integration of planning with perception, prediction, and control systems.

46openalex2026-10-09Frederik E. T. SchöllerDOI ↗

Probabilistic Ship Trajectory Forecasting using Transformer Models on AIS Data (opens in a new tab)

This study explores using Transformer and Performer models to predict ship trajectories in confined waters using AIS data. The approach incorporates positional heat maps, generated via Kernel Density Estimation and processed with a CNN, to capture historical vessel movement patterns. Models trained with quantile loss provide prediction intervals, improving collision avoidance capabilities. Results show significant performance gains over prior methods, particularly in predicting movements not evident from past positions alone.

47openalex2026-09-19Gang Chen, Jianbo Zheng, Xinhao Zhao, Zihao Wang +3 moreDOI ↗

A transformer framework for vessel trajectory prediction in complex maritime scenarios with an original vessel trajectory dataset (opens in a new tab)

This paper introduces the Segmented Autoregressive Global Attention Mechanism Transformer (SAGM-TF) to improve multi-step vessel trajectory prediction for maritime collision avoidance. The framework combines an optimized kinematic encoding module, segmented autoregressive decoding, and a cross-segment global attention mechanism to reduce error accumulation over a 90-second horizon. A new trajectory dataset, incorporating heading angle and angular velocity, was created to better capture vessel maneuvering behavior. Experimental results show the method outperforms traditional approaches in accuracy and robustness for complex maritime scenarios.

48openalex2026-04-21Takefumi Higaki, Hitoshi Yoshioka, Nobukazu Wakabayashi, Hirotada HashimotoDOI ↗

Diffusion route planner: Data-driven modeling of human ship navigation that implicitly balances safety, efficiency, and rule compliance under complex geographical constraints (opens in a new tab)

The paper introduces a diffusion route planner (DRP) that models human ship navigation using 181 days of AIS–radar-integrated data from a training ship. In tests, the DRP followed COLREGs Rules 13–17 in one-on-one scenarios and maintained safety margins in complex, multi-ship real-sea conditions. Unlike other data-driven methods, it handled realistic constraints better due to offline training on fused sensor data. The approach also showed potential to incorporate additional rules and local regulations without explicit programming.

49vocal.media2026-04-08

Autonomous Ships Market Trends: Collision Avoidance, AI Control Systems & Industry Forecast to 2033 (opens in a new tab)

The autonomous ships market is projected to grow significantly by 2033, driven by advancements in collision avoidance and AI control systems. Key trends include increased adoption of automation to improve maritime safety and efficiency. The report highlights industry forecasts, emphasizing the role of AI in enhancing navigation and operational decision-making. Market expansion is expected across commercial and defense sectors.

50openalex2026-10-02Bo Wu, Jing Yu, Zuyuan LiuDOI ↗

A Reliability-Based Robust Design Optimization Method for Unmanned Surface Vessel Berthing Trajectory Planning Under Uncertainty (opens in a new tab)

This study introduces a reliability-based robust design optimization (RBRDO) method for improving unmanned surface vessel berthing trajectory planning under uncertainty. The approach integrates line-of-sight guidance, artificial-potential-field obstacle avoidance, and sliding-mode control, while modeling collision risks for various obstacle shapes and accounting for vessel-position uncertainty. Simulations demonstrate that RBRDO outperforms baseline and deterministic optimization methods by reducing terminal distance errors, failure probability, and variability in berthing performance. The method enhances robustness and reliability in uncertain berthing conditions.

51openalex2026-09-11Shuaikun Zhang, Hui Feng, Haixiang Xu, Song Pei +1 moreDOI ↗

QSD-STFormer: Quaternion Ship Domain-guided Spatiotemporal Transformer for ship trajectory prediction (opens in a new tab)

The paper introduces QSD-STFormer, a quaternion ship domain-guided spatiotemporal transformer for ship trajectory prediction in dense maritime environments. It combines a CVAE framework with a QSD-guided Interaction Prior Module to incorporate physical navigation constraints, improving data-driven modeling. A Spatiotemporal Joint Modeling module captures dynamic ship interactions, while a GMM-Latent Representation module enhances prediction diversity. Tests on AIS data from the Port of New York and Wuhan Riverfront show a 15.28% reduction in minADE and 24.15% in minFDE compared to existing methods.

52openalex2026-09-10Doreen Jirak, Armeen Saroukanoff, Dirk van RooyDOI ↗

Understanding Operator Attitudes Toward AI-Supported Decision Making in Maritime Operations (opens in a new tab)

A survey of maritime stakeholders assessed attitudes toward AI-supported collision-avoidance systems. Participants showed general openness to maritime technology, with no age-related differences, and stable trust across scenarios, though explanation quality varied. Open responses highlighted appreciation for decision support and situation awareness but raised concerns about AI reliability and over-reliance. The study recommends designing AI systems for calibrated reliance with transparent, expert-informed features.

53openalex2026-04-28Adesegun Nurudeen Osijirin, Victor Utibe Edmond, Shamsudeen Mohammed Sada, Rafal SzlapczynskiDOI ↗

Risk-Aware Deep Learning Method for Compressing Vessel AIS Trajectories (opens in a new tab)

This paper introduces a risk-aware deep learning method to compress vessel AIS trajectory data, addressing storage, transmission, and processing challenges. The approach uses LSTM models with attention mechanisms and a risk assessment framework to prioritize high-risk segments, such as areas with vessel proximity or hazards. The method achieves a 7.5 compression ratio while retaining 95% of safety-critical features, outperforming traditional techniques. Future work will focus on real-time deployment, federated learning, and integrating additional maritime data sources.

54openalex2026-10-10Xinying Zhou, Sijin Yu, Jiaye Gong, Gongxing WuDOI ↗

Rule-Constrained Hierarchical Fusion Path Planning for Unmanned Surface Vehicles (opens in a new tab)

This study presents a hierarchical path-planning framework for unmanned surface vehicles, combining global routing and local collision avoidance while adhering to maritime rules. The method integrates adaptive A* and RRT* algorithms for global planning, alongside artificial potential fields and dynamic window approaches for local adjustments. Simulations showed 81.7% success in static planning, with lower success rates (63.3%) in integrated navigation under perturbed conditions. The approach proved feasible but highlighted inefficiencies like detours and stalled trajectories, requiring further improvements.

55openalex2026-10-01Marko Perkovič, Ivica Pavić, Maciej Gucma, Sebastian Feuerstack +1 moreDOI ↗

Maritime Security and Risk Assessments—2nd Edition: From Risk Identification to Decision Support and System Resilience (opens in a new tab)

This paper updates maritime security research, addressing challenges from digitization, automation, cyber-physical risks, and geopolitical threats. It expands on risk identification, decision-making, and system resilience, covering collision-risk sensing, human-technology interaction, and port safety. The studies highlight gaps like limited validation, incomplete causal analysis, and weak cross-scale integration. Effective implementation demands trustworthy data, uncertainty management, and coordinated response plans for cyber-physical and security failures.

56crossref2026-06-15Fulko van WestrenenDOI ↗

On the shape, size and realisation of ship safety domains on the North Sea (opens in a new tab)

This study analyzes ship safety domains in the North Sea using AIS data from 2014, focusing on vessels over 45 meters long traveling at 5–25 knots. Results confirm that safety domains are elliptical, with the major axis expanding at higher speeds while the minor axis remains constant, and ship size does not significantly affect domain dimensions. Avoidance maneuvers rely on CPA and TCPA calculations, typically initiated 600–1800 seconds before the closest point of approach, with modest course changes to maintain the elliptical domain. The findings suggest elliptical domains could improve conflict-detection algorithms and navigation systems for both crewed and autonomous ships.

57openalex2026-09-13Sijin Yu, Yunbo Li, Jiaye GongDOI ↗

A risk-aware multimodal temporal reinforcement learning approach for collision avoidance of unmanned surface vehicles in multi-dynamic-obstacle scenarios (opens in a new tab)

This paper introduces a Risk-Aware Multimodal Temporal Proximal Policy Optimization algorithm for autonomous collision avoidance in unmanned surface vehicles navigating dynamic maritime environments with varying obstacles. The method uses a collision risk index and Transformer-based network to encode risk maps and vehicle states, improving spatiotemporal feature extraction. Training is enhanced with curriculum learning, progressing from simple to complex tasks. Simulations demonstrate improved performance in success rate, safety, and adaptability compared to baseline methods.

58openalex2026-04-28Yoshiaki Kunieda, Akihiro Nunome, Emi Kanayama, Naruphun ChotechaungDOI ↗

Accident Prevention Based on Maritime Collision Cases in a Port: AI-based Measures to Prevent Ship Collision Accidents (opens in a new tab)

This study examines a 2022 ship collision in Nagoya Port caused by miscommunication via VHF radio and human error, particularly misunderstandings about vessel maneuvers at low speeds. It proposes a "digital passing agreement" system using AIS to convert verbal agreements into machine-readable data, displayed on navigation systems to improve clarity and compliance. The system detects deviations from agreed paths and integrates with Vessel Traffic Services for alerts and third-party input. The approach aims to reduce human error and improve situational awareness in port operations.

59crossref2026-04-21Ruoyun Huang, Xiang Zheng, Jianhua Wang, Gongxing Wu +2 moreDOI ↗

Nighttime Encounter Situation Recognition for Unmanned Surface Vessels Based on Images of Vessel Navigation Lights (opens in a new tab)

This study introduces an image-based method to recognize nighttime encounter situations for unmanned surface vessels (USVs) by analyzing vessel navigation lights. A 3D model of navigation lights and encounter scenarios was created following COLREGs, and a dataset was generated through simulations. The method extracts features from navigation-light images and achieves classification accuracies of 96.6%–98.3% in simulations. Field tests with a small USV demonstrated 96.2% accuracy for light recognition and 94.94% for encounter situation recognition.

60safety4sea2026-10-30

Autonomous navigation: Enhancing collision avoidance and safety at sea (opens in a new tab)

A 2026 article by safety4sea examines advancements in autonomous navigation to improve collision avoidance and maritime safety. It focuses on technologies enabling ships to operate independently while reducing human error risks. The piece highlights ongoing developments aimed at enhancing reliability and regulatory compliance in autonomous shipping. No specific methods or results are detailed in the provided text.

61openalex2026-06-17Minmin Zheng, Shuwu Wang, Haiyang Tu, Qianyu LuDOI ↗

A path planning approach for ship collision avoidance integrating BRB reasoning and velocity obstacle algorithm (opens in a new tab)

This study combines Belief-Rule-Based (BRB) reasoning and the Velocity Obstacle (VO) algorithm to improve ship collision avoidance in dense maritime traffic. BRB reasoning groups ships by movement trends and proximity, simplifying decision-making, while the VO algorithm generates safe avoidance paths. Safety domain radii for ships and groups are adjusted based on ship length. Testing in two scenarios demonstrated enhanced navigational safety and efficiency in congested waters.

62openalex2026-09-22Lulu Wang, ao xu, Mingxiao Sun, Tiantian LuanDOI ↗

Autonomous safety control of unmanned surface vessels formation driven by prescribed performance (opens in a new tab)

This paper presents a formation control framework for unmanned surface vessels (USVs) to improve cooperative tracking safety in complex marine environments. It uses a prescribed performance control method with adjustable boundaries to minimize tracking errors and reduce collision risks, while an integral disturbance observer addresses model uncertainty and external disturbances. A control barrier function (CBF) ensures safety by solving a quadratic programming problem, balancing formation tracking and security. Simulations show the method outperforms others in response speed, disturbance rejection, and input smoothness.

63openalex2026-08-19Sung‐Jo Yun, Hyogon Kim, Ji‐Wook Kwon, Young‐Ho Choi +4 moreDOI ↗

Safety-Enhanced COLREGs-Compliant Path Planning for USVs with a CBF-Based Safety Shield (opens in a new tab)

This study introduces a path planning system for Unmanned Surface Vehicles (USVs) combining Deep Reinforcement Learning (DRL) with a Control Barrier Function (CBF)-based Safety Shield to prevent collisions and comply with maritime navigation rules (COLREGs). The system uses real-time encounter classification and a CBF-QP optimization layer to filter unsafe DRL control inputs, incorporating safety radii and COLREGs constraints. A "Shielded Training" mechanism reduces reliance on the safety filter by penalizing interventions during training. Simulations show the method lowers collision and COLREGs violation rates while maintaining real-time performance.

64crossref2026-04-20Thomas MüllerDOI ↗

SafeNav: A Safety-Constrained Reinforcement Learning Framework for Autonomous Mobile Robot Navigation with Collision-Aware Path Planning (opens in a new tab)

SafeNav is a reinforcement learning framework for autonomous mobile robots that combines a collision-aware A* path planner with a PPO-based velocity controller, linked by a safety filter using Control Barrier Functions. The system extends existing methods to handle dynamic obstacles and includes safety enhancements like a guided entropy schedule and Lagrangian constraints to limit safety violations. In 1,800 simulation trials, SafeNav achieved a 95.8% navigation success rate at 20% obstacle density, a 0.48% safety violation rate, and a 31.4% reduction in path length compared to baselines. The framework aims to improve safety, efficiency, and real-world applicability in industrial environments.

65openalex2026-06-11Nikos Kougiatsos, Abhishek Dhyani, Vasso ReppaDOI ↗

Integrating rule awareness and semantic reasoning in collision-free vessel path planning (opens in a new tab)

This paper introduces an intelligent guidance system for autonomous vessels, combining traffic rule awareness and semantic reasoning to enable collision-free navigation. The framework uses qualitative traffic rule data, environmental information, and sensor inputs to make real-time path planning decisions. A finite-state machine assigns traffic roles, while a path planner calculates safe routes considering vessels and infrastructure. The approach was validated in case studies involving head-on and crossing scenarios in short-sea and inland waterways.

66crossref2026-09-22Dongqin Li, Qi Peng, Kun Zhang, Yan FengDOI ↗

SGR-YOLO: a lightweight ship detection model integrating visual modules and cross-scale fusion for complex maritime scenarios (opens in a new tab)

The SGR-YOLO model improves ship detection in complex maritime conditions by enhancing YOLOv8n with lightweight modifications. It integrates Sim-BiFPN for cross-scale feature fusion, GhostConv to reduce computational complexity, and RepViT to strengthen small-object detection. Tested on SeaShips and McShips datasets, it achieves 98.7% and 93.3% precision, respectively, while reducing parameters by up to 5.2M. The model maintains high performance in fog, occlusion, and long-distance scenarios, offering a real-time solution for resource-limited systems.

67crossref2026-06-04Zhanheng Xie, Lei Liu, Xiaosong LiDOI ↗

Prescribed-Time Trajectory Tracking and Collision Avoidance of Unmanned Surface Vehicles for Maritime Sports Assistance (opens in a new tab)

This paper presents an adaptive predefined-time sliding mode control (APTSMC) strategy for unmanned surface vehicles (USVs) to improve trajectory tracking and collision avoidance in maritime sports support. The method uses a predefined-time sliding surface and adaptive compensation to ensure predictable convergence and disturbance rejection without prior knowledge of disturbances. A velocity-modulated artificial potential field corrects paths for obstacle avoidance, adjusting repulsion based on proximity. Simulations demonstrate the framework’s effectiveness in balancing tracking accuracy and safety in dynamic environments.

68crossref2026-10-09Shigang Li, Kezhong Liu, Yanting Tong, Weiqiang WangDOI ↗

Behavior-Informed Ship Conflict Assessment in Confined Inland Waterways: A Ship-Domain Approach Using AIS Data (opens in a new tab)

This study introduces a method to assess ship conflicts in confined inland waterways using AIS data by analyzing ship behavior and maneuvers. It filters geometry-induced turning behavior and links ship-domain models to observed avoidance actions, optimizing parameters for crossing and overtaking encounters. Analysis of 10,256 trajectories from the Yangtze River revealed distinct spatial patterns: crossing conflicts clustered in two regions, while overtaking conflicts spread along the main channel. The findings offer actionable insights for improving safety management in inland waterways.

69crossref2026-09-11Jiamian Sun, Weifeng LiDOI ↗

Safety-Aware Path Planning for Unmanned Surface Vehicle: An Adaptive Hybrid Collision Avoidance Operation (opens in a new tab)

This paper introduces an adaptive path planning algorithm for unmanned surface vehicles (USVs) to improve navigation in dynamic maritime settings. The method combines an enhanced ant colony optimization (ACO) algorithm—using a pheromone penalty mechanism and path smoothing—with a modified dynamic window approach (DWA) that optimizes overshoot, selection efficiency, and adaptive weight adjustments. Simulations show the approach reduces travel length and improves collision avoidance compared to existing methods. The result is more reliable navigation in complex marine environments.

70openalex2026-04-14Min Wang, Jinwen Gao, Chenhao Li, Mei Hong +3 moreDOI ↗

Dynamic Obstacle Avoidance Algorithm for Unmanned Vessels Based on FDWA and IBA*—IGWO Fusion (opens in a new tab)

This paper introduces a dynamic obstacle-avoidance algorithm for unmanned surface vehicles (USVs) by merging Fuzzy-enhanced Dynamic Window Approach (FDWA) with an Improved Bidirectional A*–Improved Grey Wolf Optimizer (IBA*–IGWO) framework. The method improves traditional DWA by optimizing the initial heading angle and using a fuzzy controller to adjust weight coefficients in the cost function, enhancing adaptability in varying environments. The IBA*–IGWO algorithm provides a global route, with key nodes serving as local endpoints for FDWA to ensure optimal navigation and dynamic adjustments. Simulation results confirm the method’s effectiveness in avoiding unknown and dynamic obstacles while returning USVs to pre-planned routes.

71crossref2026-09-30Haoyang Yu, Xianku Zhang, Haoyun TangDOI ↗

Toughness strengthening method of course changing control for ships based on nonlinear modification (opens in a new tab)

This study introduces a cyber-resilient ship course control method using a dual-feedback switching system that alternates between positive and negative feedback to counter cyberattacks. The approach converts large-angle collision avoidance into a smaller course deviation problem and employs a second-order gain-shaping algorithm with nonlinear compensation to improve robustness. Simulations on the "Yupeng" vessel showed a 41.7% reduction in actuator amplitude, 33.2% lower rudder frequency, and 18.6% less steering energy use while maintaining course accuracy within 0.8° during attacks. Stability was confirmed through description-function analysis and the Nyquist criterion.

72openalex2026-07-02Daniel G. SchwartzDOI ↗

Path planning for unmanned naval surface vehicles (opens in a new tab)

Daniel G. Schwartz (2026) presents new methods for unmanned surface vehicles (USVs) to avoid both fixed and moving obstacles. The approach combines a global path planner—using Grassfire, a modified Grassfire, and a new Probabilistic Roadmap variant—with a local planner that adjusts routes based on obstacle movement. The local planner prioritizes routing USVs behind moving obstacles rather than alongside them. Simulations validate the method, comparing it to the D* algorithm and other dynamic path-planning systems.

73openalex2026-10-13Masoud Ebrahimi, Jackiie Samuelsson, Carolina Smigan, Abu Naser Masud +2 moreDOI ↗

From Maritime Regulations to Verifiable Properties with LLM Assistance (opens in a new tab)

This paper introduces a method to translate maritime safety regulations (COLREGs) from a structured language (Legata) into verifiable logical properties using a four-stage pipeline. The process relies on LLM agents (GPT-4o) for initial stages, with expert oversight and model checking (Rebeca) for verification. Evaluations show improved accuracy with a "skillful" LLM configuration, achieving higher scores in syntax, coverage, and logical completeness. The work provides a reusable pipeline, evaluation metric, and insights into regulatory clauses suitable for actor-based model checking.

74crossref2026-03-31Nazreen Rusli, Zulkifli Zainal Abidin, Muhammad Aiman Norazuddin, Taufik YunaharDOI ↗

An Integrated Controller System For Unmanned Surface Vehicles (USV) (opens in a new tab)

Nazreen Rusli, Zulkifli Zainal Abidin, Muhammad Aiman Norazuddin, and Taufik Yunahar (2026) present an integrated controller system for Unmanned Surface Vehicles (USVs), designed to enhance maritime operations in environmental monitoring, resource exploration, and security. The "CxSense" controller, developed by IIUM’s Centre for Unmanned Technologies and Prostrain Technologies, meets high industry standards (IPx8, ESD, and vibration tests) for robustness and reliability. Battery-powered USVs, like those using CxSense, reduce environmental impact by eliminating direct water pollutants compared to traditional fuel-powered vessels. The system aims to improve maritime safety, efficiency, and sustainability in Malaysia and neighboring regions.

75openalex2026-10-22Zheng Wang, Yongsheng Zhao, Zhiyi Wang, Xiaohong Yan +4 moreDOI ↗

Velocity-Obstacle and COLREGs-Based Safety Control for Underactuated USVs with a Fixed-Time Fuzzy Observer: Theory and Experiment (opens in a new tab)

This study presents a safety control method for underactuated unmanned surface vehicles (USVs) navigating routes with moving obstacles. It integrates velocity-obstacle techniques with COLREGs (collision avoidance regulations) and a fixed-time fuzzy observer to ensure compliance and real-time adaptability. The approach was validated through both theoretical analysis and experimental testing. Results demonstrate improved collision avoidance while maintaining waypoint-following accuracy.

76openalex2026-10-12Marko Perkovič, Ivica Pavić, Maciej Gucma, Sebastian Feuerstack +1 moreDOI ↗

Maritime security and risk assessments—2nd edition (opens in a new tab)

The second edition of *Maritime security and risk assessments* addresses evolving challenges in maritime safety, including digitization, automation, cyber-physical risks, and geopolitical threats. It expands on the first edition by exploring risk-informed decision-making, validation, and system resilience through three research streams: collision-risk sensing, human-technology interaction, and port safety. Studies use traffic data, simulations, and accident reports but face limitations like incomplete validation, sparse data, and weak cross-scale integration. Effective implementation requires trustworthy data, uncertainty management, human-centred assurance, and preparedness for cyber-physical and security failures.

77openalex2026-05-26Yuhang Zhang, Shuqi Chai, Yukang Zhang, Liusha Yang +4 more

Trust, Geometry, and Rules: A Credibility-Aware Reinforcement Learning Framework for Safe USV Navigation under Uncertainty (opens in a new tab)

This paper introduces a reinforcement learning framework for safe USV navigation that accounts for perception uncertainty and compliance with maritime collision regulations (COLREGs). The method combines credibility-weighted learning to adjust for noisy sensor data, a geometric safety shield to prevent hazardous actions, and continuous rule-aware signals to smooth navigation decisions. The approach improves training stability and reduces collisions compared to existing methods. Simulations show better performance in dynamic environments with unreliable perception systems.

A comprehensive model for assessing the risk of close approach between maritime autonomous surface ships and objectsA comprehensive model for assessing the risk of close approach between maritime autonomous surface ships and objects ↗ (opens in a new tab)

BIT* -based Path Planning for Remotely Operated Maritime Autonomous Surface Ships Under Dynamic Obstacle UncertaintyBIT* -based Path Planning for Remotely Operated Maritime Autonomous Surface Ships Under Dynamic Obstacle Uncertainty ↗ (opens in a new tab)

A neural network-based approach to collision avoidance for maritime autonomous surface shipsA neural network-based approach to collision avoidance for maritime autonomous surface ships ↗ (opens in a new tab)

A Decoupled Global-Local Planning Framework for COLREGs-Aware Maritime Autonomous Surface Ships in Coastal WatersA Decoupled Global-Local Planning Framework for COLREGs-Aware Maritime Autonomous Surface Ships in Coastal Waters ↗ (opens in a new tab)

A ship autonomous navigation decision-making method based on the OODA loop theoryA ship autonomous navigation decision-making method based on the OODA loop theory ↗ (opens in a new tab)

Autonomous vessel collision avoidance verification using geometric optimizationAutonomous vessel collision avoidance verification using geometric optimization ↗ (opens in a new tab)

Human-in-the-loop COLREGs-compliant platoon formation control of maritime autonomous surface ships via finite state machineHuman-in-the-loop COLREGs-compliant platoon formation control of maritime autonomous surface ships via finite state machine ↗ (opens in a new tab)

Machine learning-based maritime collision risk assessment and avoidance for autonomous vessels using AIS dataMachine learning-based maritime collision risk assessment and avoidance for autonomous vessels using AIS data ↗ (opens in a new tab)

From Vessel Trajectories to Safety-Critical Encounter Scenarios: A Generative AI Framework for Autonomous Ship Digital TestingFrom Vessel Trajectories to Safety-Critical Encounter Scenarios: A Generative AI Framework for Autonomous Ship Digital Testing ↗ (opens in a new tab)

PSN-PPO: A Policy-Splitting Network-Based Deep Reinforcement Learning Approach for Autonomous Navigation of Unmanned Surface VehiclesPSN-PPO: A Policy-Splitting Network-Based Deep Reinforcement Learning Approach for Autonomous Navigation of Unmanned Surface Vehicles ↗ (opens in a new tab)

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

Autonomous collision avoidance for complex multi-ship encounters based on improved dynamic window approachAutonomous collision avoidance for complex multi-ship encounters based on improved dynamic window approach ↗ (opens in a new tab)

VesselNav-Bench: A Transparent, Reproducible Benchmark for Comparing Autonomous Ship Navigation PoliciesVesselNav-Bench: A Transparent, Reproducible Benchmark for Comparing Autonomous Ship Navigation Policies ↗ (opens in a new tab)

Learning human collision avoidance behavior for autonomous ships using adaptive neuro-fuzzy inference systemLearning human collision avoidance behavior for autonomous ships using adaptive neuro-fuzzy inference system ↗ (opens in a new tab)

A Review: Learning-Based Collision Avoidance Methods for Autonomous Vessels (2021–2025)A Review: Learning-Based Collision Avoidance Methods for Autonomous Vessels (2021–2025) ↗ (opens in a new tab)

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

Adaptive collision-avoidance timing recognition for autonomous ship encounters using multi-evidence Bayesian backward evidence detectionAdaptive collision-avoidance timing recognition for autonomous ship encounters using multi-evidence Bayesian backward evidence detection ↗ (opens in a new tab)

Sandbox environment for evaluating safety-critical future autonomous ship navigationSandbox environment for evaluating safety-critical future autonomous ship navigation ↗ (opens in a new tab)

Bridging Symmetric Dynamics and Asymmetric Semantic Objectives: Runtime-Assured Predictive Safety Control for Autonomous Surface VehiclesBridging Symmetric Dynamics and Asymmetric Semantic Objectives: Runtime-Assured Predictive Safety Control for Autonomous Surface Vehicles ↗ (opens in a new tab)

A Hierarchical Logic-Guided Reinforcement Learning Framework for Interpretable Autonomous Ship Collision AvoidanceA Hierarchical Logic-Guided Reinforcement Learning Framework for Interpretable Autonomous Ship Collision Avoidance ↗ (opens in a new tab)

Automated Generation of Functionally Complete Assurance Suites for COLREGS-Compliance of Autonomous Surface VehiclesAutomated Generation of Functionally Complete Assurance Suites for COLREGS-Compliance of Autonomous Surface Vehicles ↗ (opens in a new tab)

Real-time situational awareness and collision risk evaluation for autonomous ships under sensor uncertaintyReal-time situational awareness and collision risk evaluation for autonomous ships under sensor uncertainty ↗ (opens in a new tab)

US investigators call for autonomous vessel rules after Lake Erie collisionUS investigators call for autonomous vessel rules after Lake Erie collision ↗ (opens in a new tab)

Ship collision avoidance using timespace parametrization: Robust and risk-aware planning Using SeaCharts implementationShip collision avoidance using timespace parametrization: Robust and risk-aware planning Using SeaCharts implementation ↗ (opens in a new tab)

Identification of key risk ships for MASS collision avoidance in mixed traffic through axiomatic fuzzy set-based semantic clusteringIdentification of key risk ships for MASS collision avoidance in mixed traffic through axiomatic fuzzy set-based semantic clustering ↗ (opens in a new tab)

Compliance with International Regulations for Preventing Collisions at Sea in the Context of the Introduction of Autonomous ShippingCompliance with International Regulations for Preventing Collisions at Sea in the Context of the Introduction of Autonomous Shipping ↗ (opens in a new tab)

Real-Time Obstacle Avoidance System for Autonomous Vessels Using Sensor FusionReal-Time Obstacle Avoidance System for Autonomous Vessels Using Sensor Fusion ↗ (opens in a new tab)

IoT-Enabled Autonomous Maritime Navigation in Smart Ports: A Curriculum-Guided Shared Policy Learning FrameworkIoT-Enabled Autonomous Maritime Navigation in Smart Ports: A Curriculum-Guided Shared Policy Learning Framework ↗ (opens in a new tab)

Collision Liability for Autonomous VesselsCollision Liability for Autonomous Vessels ↗ (opens in a new tab)

Real-time Collision Avoidance Considering Ship Position Uncertainty Based on AIS DataReal-time Collision Avoidance Considering Ship Position Uncertainty Based on AIS Data ↗ (opens in a new tab)

Advances in Optimized and Safe Path Planning of Marine Autonomous Surface Vehicles: A ReviewAdvances in Optimized and Safe Path Planning of Marine Autonomous Surface Vehicles: A Review ↗ (opens in a new tab)

Algorithms for determining deadlines and COLREG-compliant manoeuvres for safe navigation in multi-ship scenarios using velocity areasAlgorithms for determining deadlines and COLREG-compliant manoeuvres for safe navigation in multi-ship scenarios using velocity areas ↗ (opens in a new tab)

Safety Analysis of Collision Avoidance for Semi-Autonomous Navigation Ship Based on STPA-CREAMSafety Analysis of Collision Avoidance for Semi-Autonomous Navigation Ship Based on STPA-CREAM ↗ (opens in a new tab)

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

Intelligent anti-collision in multi-vessel scenarios: a spatial DPAD decision-making model via accident data-drivenIntelligent anti-collision in multi-vessel scenarios: a spatial DPAD decision-making model via accident data-driven ↗ (opens in a new tab)

MBSE-based development and verification of function-oriented ODDs for autonomous shipsMBSE-based development and verification of function-oriented ODDs for autonomous ships ↗ (opens in a new tab)

Development of collision-case-based testing scenarios for validating autonomous ship collision-avoidance algorithmsDevelopment of collision-case-based testing scenarios for validating autonomous ship collision-avoidance algorithms ↗ (opens in a new tab)

Interpretation and updating of regulations for preventing collision at seaInterpretation and updating of regulations for preventing collision at sea ↗ (opens in a new tab)

South Korean province seeks autonomous ship zone off Geoje for Samsung Heavy trialsSouth Korean province seeks autonomous ship zone off Geoje for Samsung Heavy trials ↗ (opens in a new tab)

A Review of Ship Path Planning: Models, Algorithms, and ChallengesA Review of Ship Path Planning: Models, Algorithms, and Challenges ↗ (opens in a new tab)

Probabilistic Ship Trajectory Forecasting using Transformer Models on AIS DataProbabilistic Ship Trajectory Forecasting using Transformer Models on AIS Data ↗ (opens in a new tab)

A transformer framework for vessel trajectory prediction in complex maritime scenarios with an original vessel trajectory datasetA transformer framework for vessel trajectory prediction in complex maritime scenarios with an original vessel trajectory dataset ↗ (opens in a new tab)

Autonomous Ships Market Trends: Collision Avoidance, AI Control Systems & Industry Forecast to 2033Autonomous Ships Market Trends: Collision Avoidance, AI Control Systems & Industry Forecast to 2033 ↗ (opens in a new tab)

A Reliability-Based Robust Design Optimization Method for Unmanned Surface Vessel Berthing Trajectory Planning Under UncertaintyA Reliability-Based Robust Design Optimization Method for Unmanned Surface Vessel Berthing Trajectory Planning Under Uncertainty ↗ (opens in a new tab)

QSD-STFormer: Quaternion Ship Domain-guided Spatiotemporal Transformer for ship trajectory predictionQSD-STFormer: Quaternion Ship Domain-guided Spatiotemporal Transformer for ship trajectory prediction ↗ (opens in a new tab)

Understanding Operator Attitudes Toward AI-Supported Decision Making in Maritime OperationsUnderstanding Operator Attitudes Toward AI-Supported Decision Making in Maritime Operations ↗ (opens in a new tab)

Risk-Aware Deep Learning Method for Compressing Vessel AIS TrajectoriesRisk-Aware Deep Learning Method for Compressing Vessel AIS Trajectories ↗ (opens in a new tab)

Rule-Constrained Hierarchical Fusion Path Planning for Unmanned Surface VehiclesRule-Constrained Hierarchical Fusion Path Planning for Unmanned Surface Vehicles ↗ (opens in a new tab)

Maritime Security and Risk Assessments—2nd Edition: From Risk Identification to Decision Support and System ResilienceMaritime Security and Risk Assessments—2nd Edition: From Risk Identification to Decision Support and System Resilience ↗ (opens in a new tab)

On the shape, size and realisation of ship safety domains on the North SeaOn the shape, size and realisation of ship safety domains on the North Sea ↗ (opens in a new tab)

Accident Prevention Based on Maritime Collision Cases in a Port: AI-based Measures to Prevent Ship Collision AccidentsAccident Prevention Based on Maritime Collision Cases in a Port: AI-based Measures to Prevent Ship Collision Accidents ↗ (opens in a new tab)

Nighttime Encounter Situation Recognition for Unmanned Surface Vessels Based on Images of Vessel Navigation LightsNighttime Encounter Situation Recognition for Unmanned Surface Vessels Based on Images of Vessel Navigation Lights ↗ (opens in a new tab)

Autonomous navigation: Enhancing collision avoidance and safety at seaAutonomous navigation: Enhancing collision avoidance and safety at sea ↗ (opens in a new tab)

A path planning approach for ship collision avoidance integrating BRB reasoning and velocity obstacle algorithmA path planning approach for ship collision avoidance integrating BRB reasoning and velocity obstacle algorithm ↗ (opens in a new tab)

Autonomous safety control of unmanned surface vessels formation driven by prescribed performanceAutonomous safety control of unmanned surface vessels formation driven by prescribed performance ↗ (opens in a new tab)

Safety-Enhanced COLREGs-Compliant Path Planning for USVs with a CBF-Based Safety ShieldSafety-Enhanced COLREGs-Compliant Path Planning for USVs with a CBF-Based Safety Shield ↗ (opens in a new tab)

SafeNav: A Safety-Constrained Reinforcement Learning Framework for Autonomous Mobile Robot Navigation with Collision-Aware Path PlanningSafeNav: A Safety-Constrained Reinforcement Learning Framework for Autonomous Mobile Robot Navigation with Collision-Aware Path Planning ↗ (opens in a new tab)

Integrating rule awareness and semantic reasoning in collision-free vessel path planningIntegrating rule awareness and semantic reasoning in collision-free vessel path planning ↗ (opens in a new tab)

SGR-YOLO: a lightweight ship detection model integrating visual modules and cross-scale fusion for complex maritime scenariosSGR-YOLO: a lightweight ship detection model integrating visual modules and cross-scale fusion for complex maritime scenarios ↗ (opens in a new tab)

Prescribed-Time Trajectory Tracking and Collision Avoidance of Unmanned Surface Vehicles for Maritime Sports AssistancePrescribed-Time Trajectory Tracking and Collision Avoidance of Unmanned Surface Vehicles for Maritime Sports Assistance ↗ (opens in a new tab)

Behavior-Informed Ship Conflict Assessment in Confined Inland Waterways: A Ship-Domain Approach Using AIS DataBehavior-Informed Ship Conflict Assessment in Confined Inland Waterways: A Ship-Domain Approach Using AIS Data ↗ (opens in a new tab)

Safety-Aware Path Planning for Unmanned Surface Vehicle: An Adaptive Hybrid Collision Avoidance OperationSafety-Aware Path Planning for Unmanned Surface Vehicle: An Adaptive Hybrid Collision Avoidance Operation ↗ (opens in a new tab)

Dynamic Obstacle Avoidance Algorithm for Unmanned Vessels Based on FDWA and IBA*—IGWO FusionDynamic Obstacle Avoidance Algorithm for Unmanned Vessels Based on FDWA and IBA*—IGWO Fusion ↗ (opens in a new tab)

Toughness strengthening method of course changing control for ships based on nonlinear modificationToughness strengthening method of course changing control for ships based on nonlinear modification ↗ (opens in a new tab)

Path planning for unmanned naval surface vehiclesPath planning for unmanned naval surface vehicles ↗ (opens in a new tab)

From Maritime Regulations to Verifiable Properties with LLM AssistanceFrom Maritime Regulations to Verifiable Properties with LLM Assistance ↗ (opens in a new tab)

An Integrated Controller System For Unmanned Surface Vehicles (USV)An Integrated Controller System For Unmanned Surface Vehicles (USV) ↗ (opens in a new tab)

Velocity-Obstacle and COLREGs-Based Safety Control for Underactuated USVs with a Fixed-Time Fuzzy Observer: Theory and ExperimentVelocity-Obstacle and COLREGs-Based Safety Control for Underactuated USVs with a Fixed-Time Fuzzy Observer: Theory and Experiment ↗ (opens in a new tab)

Maritime security and risk assessments—2nd editionMaritime security and risk assessments—2nd edition ↗ (opens in a new tab)

Trust, Geometry, and Rules: A Credibility-Aware Reinforcement Learning Framework for Safe USV Navigation under UncertaintyTrust, Geometry, and Rules: A Credibility-Aware Reinforcement Learning Framework for Safe USV Navigation under Uncertainty ↗ (opens in a new tab)