][]

Article

Comparative Analysis of Technologies for Autonomous

Aquatic Vehicle Surveillance Systems: Applicability in the

Colombian Context

Análisis comparativo de tecnologías para sistemas de

vigilancia autónoma de vehículos acuáticos: aplicabilidad

en el contexto colombiano

Iván Camilo Leiton Murcia ¹

1

Centro de Desarrollo Tecnológico Naval (CEDNAV), Cartagena, 111321, Colombia; ivan.leiton@armada.mil.co

∗

Correspondence: ivan.leiton@armada.mil.co

Citation: Leiton, I. Comparative Analysis of Technologies for Autonomous Aquatic Vehicle Surveillance Systems: Applicability in the

Colombian Context. OnBoard Knowledge Journal 2025, 1, 7. https://doi.org/10.70554/OBJK2025.v01n02.07

Received: 22/05/2025, Accepted: 18/06/2025, Published: 16/07/2025

DOI: https://doi.org/10.70554/OBJK2025.v01n02.07

Abstract: The monitoring and surveillance of aquatic vehicles has become increasingly important globally for ensuring

national security, controlling maritime and ﬂuvial trafﬁc, preventing illicit activities, and protecting sensitive ecosystems.

This study presents a comprehensive benchmarking analysis of key technologies for autonomous surveillance systems

(ASS) speciﬁcally adapted to Colombian environmental conditions. Through systematic literature review and comparative

analysis, we evaluate optical sensors (RGB and thermal cameras), radar systems, acoustic sensors, and processing

architectures (edge vs. cloud computing) under the challenging operational scenarios typical of Colombia’s diverse

aquatic environments. The research methodology encompasses four critical technological domains: detection sensors,

processing architectures, power systems, and communication technologies. Results indicate that edge computing

architectures combined with hybrid sensor conﬁgurations (optical + thermal for general surveillance, radar + PTZ

camera for high-security applications) provide optimal performance for Colombian conditions. The study concludes with

speciﬁc recommendations for large-scale deployment considering the unique geographical, climatic, and infrastructure

constraints of the Colombian territory.

Keywords: Autonomous surveillance systems; Colombian waterways; Edge computing; Maritime surveillance; Radar

systems; Thermal imaging; Vessel detection.

Resumen: La monitorización y vigilancia de vehículos acuáticos ha adquirido creciente importancia a nivel global para

garantizar la seguridad nacional, controlar el tráﬁco marítimo y ﬂuvial, prevenir actividades ilícitas y proteger ecosistemas

sensibles. Este estudio presenta un análisis comparativo integral de las tecnologías clave para sistemas de vigilancia

autónoma (SVA) especíﬁcamente adaptadas a las condiciones ambientales colombianas. Mediante revisión sistemática de

literatura y análisis comparativo, evaluamos sensores ópticos (cámaras RGB y térmicas), sistemas radar, sensores acústicos

y arquitecturas de procesamiento (edge vs. cloud computing) bajo los escenarios operativos desaﬁantes típicos de los

diversos entornos acuáticos de Colombia. Los resultados indican que las arquitecturas de edge computing combinadas con

OnBoard Knowledge Journal 2025, 1, 7.

https://revistasescuelanaval.com/obk/

Licensed by Escuela Naval de Cadetes "Almirante Padilla", COL.

This article is freely accessible and distributed under the terms and conditions

of Creative Commons Attribution (https://creativecommons.org/licenses/by/4.0/).

OnBoard Knowledge Journal 2025, 1, 7

2 of 10

conﬁguraciones híbridas de sensores proporcionan el rendimiento óptimo para las condiciones colombianas. El estudio

concluye con recomendaciones especíﬁcas para despliegue a gran escala considerando las restricciones geográﬁcas,

climáticas y de infraestructura únicas del territorio colombiano.

Palabras clave: Detección de embarcaciones; Edge computing; Imagen térmica; Sistemas de vigilancia autónoma;

Sistemas radar; Vigilancia marítima; Vías navegables colombianas.

1. Introduction

The surveillance of aquatic vehicles represents one of the most complex challenges in the ﬁeld of

contemporary national security and environmental protection. Autonomous Surveillance Systems (ASS)

have emerged as a fundamental technological solution to address the growing demand for continuous

monitoring across extensive maritime and ﬂuvial areas [5]. These systems integrate multiple sensing

technologies, advanced image processing algorithms, and communication architectures to provide real-time

vessel detection, tracking, and classiﬁcation capabilities [17].

The development of computer vision algorithms based on deep learning has revolutionized automatic

maritime object detection capabilities. Convolutional Neural Networks (CNNs) and specialized architectures

such as YOLO (You Only Look Once) have demonstrated superior performance in vessel detection within

complex maritime environments [11]. At the same time, advances in edge computing technologies have

enabled local data processing, reducing latency and dependence on network connectivity in remote locations

[13].

Colombia, due to its privileged geostrategic position, has approximately 3,208 km of coastline distributed

between the Atlantic and Paciﬁc Oceans, in addition to an extensive hydrographic network that includes

navigable rivers such as the Magdalena, Atrato, Orinoco, and Amazon. This hydrological wealth constitutes

both an economic advantage and a security challenge, particularly considering illicit activities such as drug

trafﬁcking, smuggling, and illegal ﬁshing, which use these waterways as transportation corridors.

The environmental conditions of the Colombian territory present unique challenges for the implementa-

tion of technological surveillance systems. Climatic variability, including seasons of intense rainfall, high

relative humidity (>80%), marine salinity, intense equatorial UV radiation, and complex acoustic biodiversity,

imposes speciﬁc design requirements that are not adequately addressed by generic commercial solutions [6].

The primary objective of this research is to perform a systematic comparative analysis of existing

technologies for autonomous aquatic vehicle surveillance systems, evaluating their technical performance,

operational feasibility, and economic viability within the Colombian context. By examining sensing tech-

nologies, processing architectures, power supply systems, and communication solutions, this study seeks to

provide a technically grounded reference framework to support informed decision-making for the selection

and implementation of surveillance systems adapted to the country’s geographic, climatic, and infrastructural

realities.

To guide the reader through the scope of this study, the article is structured as follows. Section 2

outlines the main contributions of this work. Section 3 describes the methodological approach adopted for

the comparative analysis. Section 4 and Section 5 examine sensing technologies and processing architectures,

respectively, while Section ?? discusses infrastructure and support systems. Section 7 presents the comparative

evaluation of the considered technologies. Section 8 reports and discusses the results obtained from the

comparative analysis. Finally, Section 9 summarizes the main ﬁndings and highlights future research

directions.

2. Contributions

This article makes the following main contributions:

i.

Provides a systematic comparative analysis of key sensing, processing, power, and communication

technologies for autonomous aquatic vehicle surveillance systems, evaluating their technical perfor-

OnBoard Knowledge Journal 2025, 1, 7

3 of 10

mance, operational robustness, and economic feasibility within the Colombian maritime and ﬂuvial

context.

ii.

Identiﬁes and characterizes the advantages and limitations of optical, thermal, radar, and acoustic

sensing technologies under Colombia’s challenging environmental conditions, demonstrating that

no single technology is sufﬁcient and that multimodal sensor integration is essential for reliable

surveillance.

iii.

iv.

Analyzes and compares edge computing and cloud computing architectures for maritime surveillance

applications, showing that edge-based processing offers superior performance for real-time detection

and autonomous operation in scenarios with limited connectivity.

Proposes optimized technological conﬁgurations for different operational scenarios in Colombia,

including standard surveillance and high-security applications, balancing detection accuracy, latency,

energy efﬁciency, and total cost of ownership.

3. Methodology

The present study is conducted under a methodological design aimed at ensuring rigor, reproducibility,

and relevance in the analysis of the technologies under consideration. To this end, a systematic approach is

adopted that combines an exhaustive review of the scientiﬁc literature with a comparative analysis of both

quantitative and qualitative nature. The methodological framework is structured into sequential phases,

ranging from the deﬁnition of the scope and evaluation criteria to the assessment of the applicability of

technological solutions within the Colombian context. This approach enables not only the identiﬁcation of

the state of the art but also the establishment of clear and objective comparison parameters that facilitate

well-founded decision-making.

3.1. Methodological Approach

The methodology employed is based on a systematic review approach combined with quantitative

and qualitative comparative analysis. The methodological process is structured into four main phases: (1)

deﬁnition of the scope and evaluation criteria, (2) systematic review of the scientiﬁc literature, (3) comparative

analysis of technologies, and (4) evaluation of their speciﬁc applicability to Colombia.

3.2. Selection and Evaluation Criteria

The comparative analysis encompasses four critical technological domains: detection sensors, processing

architectures, power supply systems, and communication technologies. For each domain, speciﬁc evaluation

metrics were established, including detection efﬁciency, total cost of ownership (TCO), operational robustness,

and applicability under Colombian conditions.

Detection sensors were evaluated considering detection range, spatial resolution, immunity to adverse

atmospheric conditions, nighttime operation capability, and energy consumption. For processing archi-

tectures, edge computing and cloud computing approaches were compared based on inference latency,

bandwidth requirements, and infrastructure complexity [19].

3.3. Information Sources

Data collection was carried out through a systematic search of academic databases, including IEEE

Xplore, ScienceDirect, Scopus, and Google Scholar. Search strings employed key terms such as “maritime

surveillance,” “vessel detection,” “edge computing for computer vision,” and “remote sensing systems.”

In addition, technical speciﬁcations from manufacturers and reports of implementations in geographically

similar contexts were analyzed.

4. Sensing and Detection Technologies

The development of advanced maritime surveillance and monitoring systems relies heavily on the

integration of multiple sensing and detection technologies, each with speciﬁc strengths and limitations under

operational conditions. These systems range from optical and thermal cameras for visual characterization of

OnBoard Knowledge Journal 2025, 1, 7

4 of 10

vessels to radar, LiDAR, and acoustic sensors capable of operating in scenarios where direct vision is limited.

The appropriate selection and combination of these technologies is critical to ensuring robust performance in

complex environments such as Colombia’s ﬂuvial and maritime ecosystems, where atmospheric, topographic,

and biological factors impose particular constraints on sensor effectiveness.

4.1. Optical Imaging Systems

Optical sensors constitute the fundamental basis of computer vision systems for maritime surveillance.

RGB cameras operate in the visible spectrum (400–700 nm), using CMOS sensors to capture color and

reﬂectance information with high spatial resolution. These data are directly processable by convolutional

neural networks for tasks such as morphological and chromatic vessel classiﬁcation [18].

Thermal cameras represent a crucial complementary technology, operating in the long-wave infrared

(LWIR, 8–14 m) to detect radiant energy emitted by objects based on their temperature. This capability enables

the identiﬁcation of characteristic thermal gradients, such as heat generated by vessel engines contrasting

with the surrounding water temperature, facilitating effective detection independent of ambient lighting

conditions [9].

The primary vulnerability of all optical systems lies in their susceptibility to adverse atmospheric

conditions. Phenomena such as Mie scattering, caused by suspended water particles during precipitation

events or fog formation, signiﬁcantly attenuate electromagnetic radiation in both the visible and infrared

spectra, thereby reducing contrast and the effective detection range [2].

4.2. Radar and LiDAR Systems

Active radar systems provide robust detection capabilities through the emission of microwave pulses

(typically in the X-band, 8–12 GHz) and the analysis of backscattered signals. This technology offers signiﬁcant

advantages in terms of detection range (up to several kilometers), continuous 24/7 operation, and resilience

to adverse weather conditions due to the microwave wavelength being considerably larger than atmospheric

water particles [10].

However, the implementation of radar systems in Colombian ecosystems faces speciﬁc challenges. In

narrow rivers with dense vegetation, radio wave propagation may be affected by multipath phenomena,

generating false echoes and increasing background noise. In maritime environments, sea clutter caused by

wave reﬂections can mask small-sized vessels [20].

4.3. Underwater Acoustic Sensors

Acoustic detection systems operate through the propagation of sound waves in aquatic media, where

acoustic waves are transmitted more efﬁciently than electromagnetic waves. Passive sensors (hydrophones)

detect characteristic acoustic signatures generated by propeller cavitation, machinery vibrations, and hydro-

dynamic ﬂow around the vessel hull [22].

The main challenge for implementation in Colombia lies in the complex bioacoustic noise spectrum

present in high-biodiversity ecosystems. Sound emissions from marine fauna (e.g., cetaceans and crustaceans)

may overlap with acoustic signatures of interest, reducing the signal-to-noise ratio and requiring adaptive

ﬁltering and classiﬁcation algorithms [7].

5. Processing and Analysis Architectures

The processing and analysis of information captured by sensing systems constitute the functional

core of maritime surveillance platforms, as these processes determine the ability to transform raw data

into actionable information. Processing architectures range from classical computer vision techniques to

deep learning algorithms that enable more accurate vessel detection and classiﬁcation in complex scenarios.

Complementarily, the deﬁnition of the computational infrastructure, whether based on edge computing or

cloud computing, establishes the balance between response speed, resource availability, and scalability of the

implemented solutions. Together, these elements form the analytical ecosystem required to ensure robust

monitoring systems adapted to Colombia’s operational conditions.

OnBoard Knowledge Journal 2025, 1, 7

5 of 10

5.1. Computer Vision Techniques

Classical computer vision algorithms, including background subtraction and frame differencing, provide

computationally efﬁcient methods for motion detection. For object tracking, Kalman ﬁlters have demonstrated

effectiveness as recursive estimators that predict future states of dynamic systems from noisy measurements,

maintaining coherent trajectories even in the presence of temporary detection failures [16].

5.2. Deep Learning Algorithms

Convolutional neural network architectures have transformed the state of the art in maritime object

detection. Detection models are generally classiﬁed into two main categories: two-stage architectures such as

Faster R-CNN, which employ region proposal networks (RPNs) followed by classiﬁcation, and one-stage

architectures such as YOLO and SSD, which treat detection as a regression problem by processing entire

images in a single pass [8].

Recent studies have shown that optimized versions of YOLO, speciﬁcally YOLOv3-Tiny, can achieve

real-time vessel detection with accuracy exceeding 87% under variable maritime conditions, making them

particularly suitable for deployment on edge computing hardware with limited resources [13].

5.3. Edge Computing vs. Cloud Computing

The choice of processing architecture critically determines the latency between data acquisition and

alert generation. Edge computing enables local processing with millisecond-level latencies, making it

essential for applications that require immediate response and autonomous operation in locations with

limited connectivity [13].

In contrast, cloud computing offers virtually unlimited processing capacity for complex algorithms,

making it more suitable for forensic analysis, model training, and large-scale analytics. The optimal selection

depends on speciﬁc requirements related to latency, connectivity availability, and the computational resources

needed [12].

6. Infrastructure and Support Systems

The effectiveness of maritime surveillance systems depends not only on sensing and processing tech-

nologies but also on the infrastructure and support systems that ensure sustained operation in demanding

environments. These components include autonomous energy solutions capable of maintaining continuous

operation in remote locations, environmental protection mechanisms that safeguard equipment against

severe conditions of humidity, salinity, and biofouling, and communication platforms that ensure reliable

data transmission in areas with limited connectivity. The integration of these elements is fundamental to

achieving robust and resilient systems adapted to Colombia’s geographic and climatic conditions.

6.1. Power Supply Systems

Autonomous operation in remote locations without access to conventional power grids requires on-

site energy generation and storage solutions. Photovoltaic systems represent the most mature technology,

combining solar panels, charge controllers, and battery banks to ensure continuous operation [3].

In coastal regions with wind potential, hybrid microgrids that combine solar and wind energy can

signiﬁcantly increase supply reliability. Battery bank sizing must consider not only nighttime operation

but also extended periods of low solar irradiance during rainy seasons characteristic of Colombia’s tropical

climate [14].

6.2. Environmental Protection of Equipment

Electronic components require robust protection against severe environmental conditions, including

humidity, salinity, and accelerated corrosion. IP66 and IP67 protection standards represent the minimum

requirements for operation in maritime environments. The selection of corrosion-resistant materials, such as

316L stainless steel or aluminum with marine-grade coatings, is critical to ensuring operational durability [1].

OnBoard Knowledge Journal 2025, 1, 7

6 of 10

Additional considerations include protection against biofouling (the growth of marine organisms), which

can obstruct optical sensors and ventilation systems, requiring speciﬁc design strategies and preventive

maintenance protocols [4].

6.3. Communication Technologies

Data transmission from remote locations presents signiﬁcant challenges due to infrastructure limitations.

Cellular networks (4G/5G) offer high bandwidth but limited coverage in rural and maritime areas. Satellite

communication provides global coverage but entails high operational costs and increased latency [21].

LoRaWAN technologies represent a promising alternative for low-bandwidth data transmission (e.g.,

alerts and metadata), offering extended range and low energy consumption, particularly when integrated

with Low Earth Orbit (LEO) satellite constellations for areas without terrestrial coverage [15].

7. Comparative Analysis of Technologies

The comparative analysis of technologies constitutes a key stage in identifying strengths, limitations,

and potential synergies among different solutions applicable to maritime and ﬂuvial surveillance. Based

on technical and operational criteria, detection sensors, processing architectures, and support systems are

evaluated in order to determine which combinations are most viable within the Colombian context. This

approach enables not only the characterization of the individual performance of each technology but also the

recognition of the need for multimodal integration and complementary infrastructures to ensure continuous,

accurate, and efﬁcient operation under the country’s speciﬁc environmental, geographic, and logistical

conditions.

7.1. Comparison of Detection Sensors

As presented in Table 1, the comparison of sensors demonstrates that no single technology is capable of

comprehensively meeting the requirements of maritime and ﬂuvial surveillance in Colombia. RGB cameras

provide high visual resolution under favorable lighting conditions but are ineffective in darkness or fog,

highlighting the complementary role of thermal cameras in nighttime operations. Radar systems, in turn,

ensure long-range detection under all weather conditions, although they face limitations in scenarios with

intense wave activity, while acoustic sensors are effective in discrete underwater environments, with the

challenge of mitigating interference from bioacoustic noise. Overall, the ﬁndings summarized in Table 1 lead

to the conclusion that multimodal integration of these technologies represents the most suitable strategy to

address the diversity of environmental and operational conditions present in the country’s maritime and

ﬂuvial ecosystems.

7.2. Evaluation of Processing Architectures

As shown in Table 2, each processing and analysis architecture presents advantages and limitations that

condition its applicability in maritime and ﬂuvial surveillance environments. Classical computer vision offers

low-cost solutions with low computational complexity, although it is restricted to controlled scenarios. In

contrast, deep learning models provide high accuracy and strong generalization capabilities, at the expense of

requiring large volumes of labeled data and greater computational resources. Regarding infrastructure, edge

computing stands out for its low latency and operational autonomy under limited connectivity conditions,

making it ideal for real-time alert generation, while cloud computing offers virtually unlimited processing

capacity and is more suitable for forensic analysis and complex model training. Collectively, Table 2 shows

that the strategic integration of these architectures is essential to balance accuracy, response speed, and

scalability in monitoring systems tailored to Colombian needs.

7.3. Analysis of Support Systems

As summarized in Table 3, power supply and communication systems represent critical components for

ensuring the sustained operation of surveillance platforms in maritime and ﬂuvial environments. Photovoltaic

solutions stand out due to their technological maturity and decreasing costs, although they face limitations

OnBoard Knowledge Journal 2025, 1, 7

7 of 10

Table 1. Comparison of Detection Technologies Applicable to Maritime and Fluvial Surveillance in Colombia

Technology

Operating Principle

Main Advantage

Critical Limitation

Applicability

Colombia

in

RGB Camera

Capture of reﬂected High spatial resolu- Inoperative in dark- Optimal for ports

light in the visible tion for detailed vi- ness; severe degrada- and marinas with

spectrum

nm)

(400–700 sual identiﬁcation

tion under rain or fog adequate

conditions

lighting

Thermal Cam- Detection of infrared 24/7 operation inde- Limited resolution; Excellent for night-

era

radiation emitted by pendent of ambient susceptible to heavy time surveillance in

objects (8–14 µm)

lighting

rainfall

rivers and coastal ar-

eas

Radar

Emission and recep- Long-range detection Lack of visual identiﬁ- Ideal for trafﬁc con-

tion of microwave sig- under all weather con- cation; sea clutter due trol in main maritime

nals (8–12 GHz)

Acoustic Sen- Detection of sound Discreet underwater Interference from ma- Effective in narrow

sor waves propagating detection without line rine fauna bioacoustic channels with low

underwater of sight noise biodiversity levels

Source: The authors.

ditions

to waves

and ﬂuvial channels

Table 2. Comparison of Processing and Analysis Architectures for Maritime and Fluvial Surveillance Applications in

Colombia

Architecture

Methodology

Main Requirement

Operational Advan- Recommended Ap-

tage plication

Classical Vi- Explicit algorithms Controlled

scenes Low computational Basic motion detec-

back- cost; no training re- tion

quired

sion

(background

traction,

sub- with

Kalman grounds

static

ﬁltering)

CNN / Deep Data-driven

Learning

mod- Large datasets of la- High accuracy and Accurate classiﬁca-

strong generalization tion of vessel types

capability

els (YOLO, Faster beled images

R-CNN)

Edge Comput- Local processing on Hardware with in- Minimal latency; au- Real-time alerts with

ing

embedded devices

ference

(GPU/TPU)

Cloud Com- Remote processing on Reliable

capability tonomous operation

minimal bandwidth

usage

high- Virtually unlimited Forensic analysis and

con- computational capac- model training

ity

puting

centralized servers

bandwidth

nectivity

Source: The authors.

during rainy seasons. Wind and micro-hydropower can complement energy generation, with constraints

related to local resource availability and maintenance requirements. Regarding communications, cellular

networks offer high speed and low latency but limited coverage in remote areas, while satellite systems

ensure global reach at signiﬁcantly higher costs. Overall, Table 3 shows that the optimal selection of support

infrastructure requires a balance among reliability, cost, and adaptability to Colombia’s speciﬁc geographic

and climatic conditions.

8. Results

The deﬁnition of optimal technological conﬁgurations is essential for adapting maritime and ﬂuvial

surveillance systems to Colombia’s particular conditions. Based on the results of the comparative analysis,

integrated schemes are proposed that seek to balance performance, cost, and resilience against the country’s

environmental, geographic, and operational challenges. These conﬁgurations consider not only the selection

of appropriate sensors and processing architectures but also energy infrastructure requirements, environ-

OnBoard Knowledge Journal 2025, 1, 7

8 of 10

Table 3. Comparison of Power Supply and Communication Technologies for Maritime and Fluvial Surveillance Systems

in Colombia

Component

Technology

Main Advantage

Design Considera- Challenge in Colom-

tion bia

Power Supply

Solar Photovoltaic

Technological matu- Battery sizing to en- Low solar irradiance

rity and decreasing sure system auton- during rainy seasons

costs

Complementary 24/7 Assessment of local Limited applicability

power generation wind resources to coastal regions

Stable generation in Minimum ﬂow rate Complex mainte-

rivers with constant and head require- nance and potential

ﬂow ments ecological impact

omy

Wind (Micro)

Micro-hydropower

Communication Cellular (4G/5G)

Satellite (LEO/GEO)

High bandwidth and On-site veriﬁcation of Limited coverage in

low latency

network coverage

remote areas

Global coverage

Cost–latency trade- High cost per trans-

off depending on mitted gigabyte

application

Source: The authors.

mental protection, and communication systems, in order to ensure sustainable and effective solutions across

different application scenarios.

8.1. Optimal Conﬁgurations for Colombia

Based on the comparative analysis, two optimal technological conﬁgurations are identiﬁed for different

operational scenarios in Colombia:

Standard Conﬁguration: For general surveillance and trafﬁc control on major rivers, the combination of

an RGB camera + thermal camera + edge computing processing provides the best cost–beneﬁt balance. This

conﬁguration ensures 24/7 detection with visual classiﬁcation capabilities, operating autonomously with

minimal latency and moderate bandwidth requirements.

High-Security Conﬁguration: For border surveillance and critical infrastructure protection, the integra-

tion of radar + PTZ camera + hybrid edge computing offers superior capabilities. Radar provides reliable

detection under adverse weather conditions, while the PTZ camera enables precise identiﬁcation of detected

targets.

9. Conclusions

This study has demonstrated that there is no universally optimal technological solution for aquatic

vehicle surveillance; appropriate selection requires a careful balance among detection performance, envi-

ronmental robustness, and total cost of ownership. In the Colombian context, edge computing architectures

combined with hybrid sensor conﬁgurations (optical–thermal or radar–optical) provide the most robust

overall performance.

The main challenges identiﬁed include energy management during extended rainy periods, protection

against accelerated corrosion in saline environments, and mitigation of bioacoustic interference in high-

biodiversity ecosystems. These challenges require tailored engineering solutions that account for the speciﬁc

characteristics of the national territory.

Future work should focus on the development of adaptive sensor fusion algorithms capable of automat-

ically optimizing the contribution of each detection modality based on real-time environmental conditions.

Additionally, further research is needed on acoustic classiﬁcation algorithms speciﬁcally designed to distin-

guish between vessels of interest and marine fauna in tropical ecosystems.

OnBoard Knowledge Journal 2025, 1, 7

9 of 10

The large-scale deployment of surveillance networks will require the development of optimized commu-

nication protocols for efﬁcient transmission of critical data under bandwidth constraints, as well as predictive

maintenance strategies based on data analytics from integrated environmental sensors.

Author Contributions: Iván Leiton: Conceptualization, Methodology, Software, Validation, Formal analysis, Investiga-

tion, Resources, Data curation, Writing – original draft, Writing – review editing, Visualization, Supervision, Project

administration, Funding acquisition.

All authors have read and agreed to the published version of the manuscript. Refer to the taxonomía CRediT for term

explanations. Authorship should be limited to those who have contributed substantially to the work reported.

Funding: This research received no external funding.

Institutional Review Board Statement: Not applicable, since the present study does not involvehuman personnel or

animals.

Informed Consent Statement: This study is limited to the use of technological resources, so nohuman personnel or

animals are involved.

Conﬂicts of Interest: Under the authorship of this research, it is declared that there is no conﬂict of interest with the

present research.

References

1. Adamo, F., Attivissimo, F., Carducci, C. G. C., and Lanzolla, A. M. L. (2014). A smart sensor network for sea water

quality monitoring. IEEE Sensors Journal, 15(5):2514–2522.

2. AlMansoori, A. A. S., Swamidoss, I., and Niwas, S. I. (2020). Analysis of different tracking algorithms applied

on thermal infrared imagery for maritime surveillance systems. In Artiﬁcial Intelligence and Machine Learning for

Multi-Domain Operations Applications II, volume 11543.

3. Copping, A., LiVecchi, A., Spence, H., et al. (2018). Maritime renewable energy markets: Power from the sea. Marine

Technology Society Journal, 52(5):99–109.

4. Delauney, L., Compere, C., and Lehaitre, M. (2010). Biofouling protection for marine environmental sensors. Ocean

Science, 6(2):503–511.

5. Gamage, C., Dinalankara, R., Samarabandu, J., et al. (2023). A comprehensive survey on the applications of machine

learning techniques on maritime surveillance to detect abnormal maritime vessel behaviors. WMU Journal of Maritime

Affairs, 22(3):343–367.

6. Guo, Y., Lu, Y., and Liu, R. W. (2022). Lightweight deep network-enabled real-time low-visibility enhancement for

promoting vessel detection in maritime video surveillance. The Journal of Navigation, 75(2):230–250.

7. Hastie, G. D., Wu, G. M., Moss, S., Jepp, P., et al. (2019). Automated detection and tracking of marine mammals: A

novel sonar tool for monitoring effects of marine industry. Aquatic Conservation: Marine and Freshwater Ecosystems,

29(6):1019–1030.

8. He, G., Wang, W., Shi, B., Liu, S., Xiang, H., et al. (2022). An improved yolo v4 algorithm-based object detection

method for maritime vessels. International Journal of Software Engineering and Applications, 11(4):23–35.

9. Helgesen, H. H., Leira, F. S., Fossen, T. I., and Johansen, T. A. (2018). Tracking of ocean surface objects from unmanned

aerial vehicles with a pan/tilt unit using a thermal camera. Journal of Intelligent & Robotic Systems, 91(3–4):775–793.

10. Helgesen, Ø. K., Brekke, E. F., Stahl, A., and Engelhardtsen, Ø. (2020). Low altitude georeferencing for imaging

sensors in maritime tracking. IFAC-PapersOnLine, 53(2):14256–14261.

11. Li, H., Deng, L., Yang, C., Liu, J., and Gu, Z. (2021). Enhanced yolo v3 tiny network for real-time ship detection from

visual image. IEEE Access, 9:16890–16900.

12. Li, Z., Liu, T., Li, S., You, L., Liang, S., et al. (2024). An unmanned trafﬁc command system for controlled waterway in

inland river: An edge-centric iot approach. Unmanned Systems, 12(3):445–462.

13. Liu, R. W., Guo, Y., Nie, J., Hu, Q., Xiong, Z., et al. (2022). Intelligent edge-enabled efﬁcient multi-source data

fusion for autonomous surface vehicles in maritime internet of things. IEEE Transactions on Green Communications and

Networking, 6(2):1054–1064.

14. Omondi, J. C. (2017). Improving maritime surveillance in kenya’s remote coastal islands: Application of renewable

energy solutions. Master’s thesis, World Maritime University, Malmö, Sweden.

15. Pensieri, S., Viti, F., Moser, G., Serpico, S. B., and Bordone, A. (2021). Evaluating lorawan connectivity in a marine

scenario. Journal of Marine Science and Engineering, 9(11):1218.

OnBoard Knowledge Journal 2025, 1, 7

10 of 10

16. Perera, L. P., Oliveira, P., and Soares, C. G. (2012). Maritime trafﬁc monitoring based on vessel detection, tracking,

state estimation, and trajectory prediction. IEEE Transactions on Intelligent Transportation Systems, 13(3):1188–1200.

17. Rekavandi, A. M., Xu, L., Boussaid, F., et al. (2025). A guide to image- and video-based small object detection

using deep learning: Case study of maritime surveillance. IEEE Transactions on Intelligent Transportation Systems,

26(1):123–145.

18. Santos, V. G., Pereira, D. S., Nascimento, L. B. P., et al. (2022). Cnn-based boat detection for environmental protection

area monitoring. In Proceedings of the Congresso Brasileiro de Automática, pages 1–8.

19. Sathiyamurthy, A. and Naidu, S. G. S. (2025). Designing edge computing solutions for real-time vessel tracking and

collision avoidance. Journal of Wireless Mobile Networks, Ubiquitous Computing, and Dependable Applications, 16(2):45–62.

20. Shao, Z., Lyu, H., Yin, Y., Cheng, T., Gao, X., et al. (2022). Multi-scale object detection model for autonomous ship

navigation in maritime environment. Journal of Marine Science and Engineering, 10(11):1783.

21. Ullah, M. A., Mikhaylov, K., and Alves, H. (2021). Enabling mmtc in remote areas: Lorawan and leo satellite

integration for offshore wind farm monitoring. IEEE Transactions on Communications, 69(12):8744–8758.

22. Zhang, C. and Xiao, F. (2021). Overview of data acquisition technology in underwater acoustic detection. Procedia

Computer Science, 183:509–516.

Authors’ Biography

Iván Leiton Electronics engineer, graduated from the University of Ibagué, with experience in

mechatronics, machine vision, artiﬁcial intelligence, and control engineering, areas in which

he has developed highly demanding technical projects. He completed a Master’s degree in

Renewable Energies at the Universidad Internacional de La Rioja (UNIR). His ﬁrst professional

experience was in a multinational company within the energy sector, while simultaneously

participating in the design of printed circuit boards for research projects at the Universidad

Cooperativa de Colombia, focused on Internet of Things (IoT) solutions for agriculture. During

this period, he strengthened his programming skills and advanced artiﬁcial intelligence tech-

niques. On July 8, 2024, he entered the Escuela Naval de Cadetes “Almirante Padilla” with

the ﬁrm purpose of putting his knowledge at the service of the Colombian Navy. After being

commissioned as an ofﬁcer in December 2024, he was assigned to the Centro de Desarrollo

Tecnológico Naval, where he actively participates in research, innovation, and technological

development projects aimed at strengthening the institution’s strategic capabilities.

Disclaimer/Editor’s Note: Statements, opinions, and data contained in all publications are solely those of the individual

authors and contributors and not of the OnBoard Knowledge Journal and/or the editor(s), disclaiming any responsibility

for any injury to persons or property resulting from any ideas, methods, instructions, or products referred to in the

content.