Unmanned Surface Vehicle Research Articles (Page 1)

For efficient and safe navigation for an autonomous surface vehicle (ASV), this paper proposes an autonomous navigation behavior framework that integrates deep reinforcement learning (DRL) to achieve autonomous decision-making and low-level control actions in path following and collision avoidance. By controlling both the propeller speed and the rudder angle, the policy of each behavior pattern is trained with the soft actor–critic (SAC) algorithm. Moreover, a dynamic obstacle trajectory predictor based on the Kalman filter and the long short-term memory module is developed for obstacle avoidance. Simulations and physical experiments using an under-actuated very large crude carrier (VLCC) model indicate that our DRL-based method produces appreciable performance gains in ASV autonomous navigation under environmental disturbances, which enables forecasting of the expected state of a vessel over a future time and improves the operational efficiency of the navigation process.

Abstract This study proposes a dynamic in situ counting method to overcome the challenges of low efficiency and high costs associated with manual counting in Holothurian aquaculture. The method is based on an enhanced YOLOv11n detector and ByteTrack framework, aiming to achieve high-precision automatic counting in complex underwater environments. First, the feature extraction capabilities of the YOLOv11n detector are enhanced through the integration of the SENetV2 attention mechanism, and the WIoU loss function is employed to optimize bounding box regression, significantly improving target detection performance. Secondly, to resolve issues related to camera non-linear motion and target occlusion, the ByteTrack tracker integrates global motion compensation (GMC) technology, and a trajectory interpolation post-processing strategy is developed to enhance the stability of target IDs. Experimental results demonstrate that the improved model achieves an accuracy of 93.60%, recall of 91.70%, and mAP@0.5 of 93.30% for detection tasks. In multi-object tracking tasks, the high-order tracking accuracy (HOTA), multi-object tracking accuracy (MOTA), and IDF1 score improved to 74.43%, 80.95%, and 89.35%, respectively. Testing in real aquaculture environments shows that the model achieves a counting accuracy of 96.04% and a root mean square error of 1.58, significantly outperforming existing mainstream methods. This research presents an efficient and reliable solution for the dynamic monitoring of underwater Holothurian farming, providing substantial application value for smart aquaculture management and decision support.

Abstract. The ocean skin layer, which covers the upper millimetre of the sea surface, regulates the exchange of heat, gases, and freshwater between the atmosphere and the ocean. However, there is a lack of small-scale mechanistic understanding of these fluxes, especially under abrupt meteorological shifts, due to observational challenges during stormy conditions in the open sea. This study provides unique data on temperature and salinity anomalies between the skin layer and a depth of 100 cm during atmospheric cold pools, which induce abrupt shifts in air temperature, wind speed, precipitation, and heat fluxes. We determined how these abrupt meteorological shifts forced the anomalies and altered the conditions at the air–sea boundary layer during three events monitored by an autonomous surface vehicle. Two cold pool events were observed in the harbour of Bremerhaven and one event in the North Sea. Here, we show that the skin layer instantly reacts to abrupt meteorological shifts forced by cold pools. The average temperature change in the skin layer was twice as much as at a depth of 100 cm. An abrupt change in meteorological conditions, shifting the net heat flux from positive to negative, can turn a warm skin layer into a cooler layer compared with the 100 cm depth. Salinity anomalies in the harbour were less affected by abrupt meteorological shifts, including freshwater fluxes, than those in the North Sea event. The current velocities showed that changes in wind direction could alter the surface current direction, and that the backscatter signal consistently reflects wind-induced mixing, with higher backscatter observed during increased wind conditions. This study reveals the complex relationships between atmospheric conditions and oceanic responses and provides valuable information for understanding air–sea interactions and their implications for climate dynamics.

Global road safety reports identify human factors as the leading causes of traffic accidents, particularly behaviors such as speeding, drunk driving, and driver distraction, emphasizing the need for autonomous driving technologies to enhance transport safety. This research aims to provide a practical model for the development of autonomous driving systems as part of an autonomous transportation system for inter-building passenger mobility, intended to enable safe and efficient short-distance transport between buildings in semi-open environments such as university campuses. This work presents a fully integrated autonomous platform combining LiDAR, cameras, and IMU sensors for mapping, perception, localization, and control within a drive-by-wire framework, achieving superior coordination in driving, braking, and obstacle avoidance and validated under real campus conditions. The electric golf cart prototype achieved centimeter-level mapping accuracy (0.32 m), precise localization (0.08 m), and 2D object detection with an mAP value exceeding 70%, demonstrating accurate perception and positioning under real-world conditions. These results confirm its reliable performance and suitability for practical autonomous operation. Field tests showed that the vehicle maintained appropriate speeds and path curvature while performing effective obstacle avoidance. The findings highlight the system’s potential to improve safety and reliability in short-distance autonomous mobility while supporting scalable smart mobility development.

Introduction As cyber-physical systems become increasingly virtualized, digital twins have emerged as essential components for real-time monitoring, simulation, and control. However, their growing complexity and exposure to dynamic network environments make them vulnerable to sophisticated cyber threats. Traditional rule-based and machine-learning-based security models often fail to adapt in real time to evolving attack patterns, particularly in decentralized and resource-constrained settings. Methods This study introduces the Neuromorphic Cyber-Twin (NCT), a brain-inspired architectural framework that integrates spiking neural networks (SNNs) and event-driven cognition to enhance adaptive cyber defense. The NCT leverages neuromorphic principles such as sparse coding, temporal encoding, and spike-timing-dependent plasticity (STDP) to transform telemetry data from the digital-twin layer into spike-based sensory inputs. A layered cognitive architecture continuously monitors behavioral deviations, infers anomalies, and autonomously adapts its defensive responses in alignment with system dynamics. Results Lightweight prototype simulations demonstrate the feasibility of NCT-based event-driven anomaly detection and adaptive defense. The results highlight advantages in low-latency detection, contextual awareness, and energy efficiency compared with conventional machine-learning models. Discussion The NCT framework represents a biologically inspired paradigm for scalable, self-evolving cybersecurity in virtualized ecosystems. Potential applications include infrastructure monitoring, autonomous transportation, and industrial control systems. Comprehensive benchmarking and large-scale validation are identified as future research directions.

This paper presents a global fixed-time control framework to address the target circumnavigation tracking problem of underactuated unmanned surface vehicles (USV) under unknown velocity states, lumped uncertainties, and actuator saturation. At the core of this approach is a novel fixed-time target-enclosing line-of-sight (FTTELOS) guidance law, designed to generate the desired heading angle and surge velocity. To estimate unknown velocities, external disturbances, and unmeasured system states, a set of fixed-time observers is constructed, consisting of a velocity observer, a disturbance observer, and a high-dimensional extended state observer (HFTESO). Moreover, to enhance robustness and effectively tackle actuator saturation, the control scheme incorporates a fixed-time sliding mode controller, a dynamic auxiliary system, and a fixed-threshold event-triggered mechanism. Simulation results using SimuNPS demonstrate that the proposed method enables rapid and smooth target circumnavigation, with all system errors converging to an arbitrarily small neighborhood of the origin within a fixed time. Theoretical analysis and simulation studies confirm the effectiveness and robustness of both the FTTELOS guidance law and the integrated control strategy. Quantitatively, compared with the traditional target-enclosing line-of-sight (TELOS) method, the proposed FTTELOS reduces the convergence time of the distance error δe from 13.64 s to 10.22 s and the angular error ϕe from 10.46 s to 7.52 s, demonstrating a significant improvement in convergence speed and overall control performance.

Introduction: Technological developments in the maritime sector have led to innovations such as unmanned surface vessels (Maritime Autonomous Surface Ships/MASS). The emergence of MASS brings efficiency, safety, and new innovations to the world of shipping, but it also poses legal challenges, particularly regarding the application of international regulations that have traditionally governed manned vessels.Purposes of the Research: This study aims to analyze the application of legal provisions and identify liability mechanisms for MASS under international law, as well as compare practices across several countries.Methods of the Research: The methodology employed is normative legal research using a legislative, comparative legal, and conceptual approach, utilizing primary legal sources such as international conventions (UNCLOS, SOLAS, and IMO regulations) and relevant literature.Results of the Research: The results of the study indicate that most international legal instruments have not yet fully accommodated the characteristics and regulatory needs of autonomous ships, particularly in terms of the definition of legal subjects, the role of the captain, and accountability mechanisms in the event of an incident. This is because most of these international legal provisions are still based on the assumption that ships are controlled by humans. Some countries, such as the United Kingdom, Norway, and the United States, have begun to formulate specific regulations to govern MASS that can fill this gap. Therefore, accountability is needed, which indicates the need for updating and harmonizing international rules to address the challenges arising from technological developments in the maritime sector. Additionally, it is important to develop national implementation guidelines aligned with the principles of international maritime law to ensure maritime safety, marine environmental protection, and legal certainty.

Submesoscale processes, characterized by strong vertical velocities that generate sea surface temperature (SST) fronts as well as O(1) Rossby number (Ro), are critical to ocean mixing and biogeochemical transport, yet their observation is hampered by cost and spatial limitations. Hence, this study proposes an adaptive sampling framework for unmanned surface vehicles (USVs) that integrates Gaussian process regression (GPR) with submesoscale physical characteristics for efficient, targeted sampling. Three composite-kernel GPR models are developed to predict SST, zonal velocity U, and meridional velocity V, providing predictive fields to support adaptive path planning. A robust coupled gradient indicator (CGI) is further introduced to identify SST frontal zones, where the maximum CGI values are used to select candidate waypoints. Connecting these waypoints yields adaptive paths aligned with frontal structures, while a Ro threshold (0.5–2) automatically triggers spiral-intensive sampling to collect more useful data. Simulation results show that the planned paths effectively capture SST gradient and submesoscale dynamics. The final environment reconstruction achieved the desired accuracy after model retraining, and deployment analysis informs optimal platform deployment. Overall, the proposed framework couples environmental prediction, adaptive path planning, and intelligent sampling, offering an effective strategy for advancing the observation of submesoscale ocean processes.

This study presents a DSDEVS-based method to accelerate simulation execution for AI training in USV (Unmanned Surface vehicle) naval combat scenarios. The proposed approach introduces an event filtering technique that selectively suppresses low-importance sensing events based on the distance to enemy targets. By dynamically adjusting structural couplings and modifying sensing frequency through domain-specific thresholds, the method reduces execution time while maintaining a balance between speed and fidelity. Two key parameters—Event Filtering Distance (EFD) and Sensor Acceleration Time Advance (SATA)—enable conditional event filtering and time advance adjustments within the sensor model. Experimental results demonstrate a 3.03 improvement in runtime, highlighting the effectiveness of the method and the trade-off between simulation speedup and fidelity.

Autonomous vessels driven by renewable energy are increasingly envisioned as vital for sustainable ocean?operations such as environmental monitoring, offshore power generation, and long-haul unmanned surface vehicles. Implementing fine-scale control of these systems has proven challenging however,?due to time-varying sea-state dynamics, sporadic energy inputs, the possibility of failure at the component level, and the requirement for coordination between multiple agents. In the article, an end-to-end deep reinforcement learning-based hierarchical control solution with real-time navigation and?its synthesis for energy optimization is proposed. It combines high-level energy regulation with low-level actuator scheduling so as to react to the variations of?the environment and internal perturbations. Simulations using actual wave realizations, sensor failures, actuator outages, and network communication variation were used?to demonstrate the performance of the control system in the following 5 performance aspects: energy saving, navigation accuracy, communication reliability, fault tolerant and multi-agent coordination. Results indicate that the architecture sustained over 80% of the performance and achieved energy efficiencies up to 54.5% in the?best case under failure scenarios. Performance-measures demonstrated reasonable scalability?up to 5–7 agents without significant communication overhead. The findings support the applicability of deep reinforcement learning for real-time maritime control under uncertainty, offering a viable alternative to conventional rule-based or predictive control strategies. The framework’s modular design allows for future integration with federated learning, hybrid control models, or autonomous deployment. The article contributes to the growing field of intelligent marine systems by providing a robust and adaptable control strategy for sustainable and scalable operations in autonomous maritime environments.

Anomaly detection in unmanned surface vehicles via multimodal learning

Evaluating the effect of risk metrics for supporting operational decision-making by autonomous surface vehicles

Finite-time output feedback control of unmanned marine surface vessels with rotatable thrusters and propellers under deception attacks in an IoT environment

Bayesian deep learning based semantic segmentation for unmanned surface vehicles in uncertain marine environments

Efficiency advantage actor-critic reinforcement learning control for an unmanned surface vehicle with unknown uncertainties

An overview of Unmanned Surface Vehicles: Methods, practices, and applications

Energy-efficient tracking control of unmanned surface vehicles: A fixed-time pose and velocity constrained strategy

Parameter identification and real-time motion prediction for a water-jet unmanned surface vehicle based on online sparse least squares support vector machine algorithm

Cooperative game method of heterogeneous unmanned surface vehicles based on distributed decision-making framework

Observer-based adaptive tracking control for unmanned surface vehicles under actuator saturation in high sea states