The Future of Unmanned Naval Systems:
An Overview Of Efficiency Enhancements
and Warfare Upgrades for Maritime
Drones
University of Turku
Department of Computing
Master of Science in Technology Thesis
Robotics and Autonomous Systems
June 2025
Jere Leman
Supervisors:
Paavo Nevalainen
The originality of this thesis has been checked in accordance with the University of Turku quality assurance
system using the Turnitin OriginalityCheck service.
UNIVERSITY OF TURKU
Department of Computing
Jere Leman: The Future of Unmanned Naval Systems: An Overview Of Efficiency
Enhancements and Warfare Upgrades for Maritime Drones
Master of Science in Technology Thesis, 75 p.
Robotics and Autonomous Systems
June 2025
The dramatic rise of tensions between various countries has been escalating since
the larger-scale Russian invasion of Ukraine in 2022. At the same time, the robotics
and autonomous systems fields are blooming with new inventions and technology for
both civilian and military applications. Most of the existing autonomous systems
are built on ground, air, or space platforms. However, in this thesis, maritime drones
will be explored comprehensively to improve their efficiency and military aspects,
considering that the field of marine robotics lags behind other robotics fields. The
development of autonomous weapon systems is highly connected to the advance-
ments in AI technology, which allows the fundamentals for such maritime drone sys-
tems. The two largest competitors, the United States of America and China, bolster
billions annually to fund research and development of maritime drones. However,
Ukraine and its innovative solutions offer more concrete use cases and developments
that have come to the public’s attention during their asymmetric warfare against
Russia. Ukraine has effectively defended its territorial waters using maritime drones
alongside its much smaller navy compared to the Russian naval fleet.
The reasoning for this thesis is to address three research questions:
1. What are the strategic and tactical implications of maritime drones
in modern warfare?
2. How can maritime drones be optimized and equipped better?
3. What are the ethical and environmental considerations surrounding
the deployment of maritime drones? And what kind of regulatory
measures have been set for maritime drones?
The thesis will present various maritime drones and their properties. Additionally, it
will feature a maritime drone developed in collaboration with the Finnish National
Defence University, the University of Turku, and the Turku University of Applied
Sciences. The Meri-Turso is a maritime drone that improves efficiency and afford-
ability through innovative design. The role of naval drones in modern warfare will
be examined through an analysis of current conflicts.
Keywords: Maritime, Naval, USV, UUV, Autonomous Weapon Systems, Warfare
0.1 Acknowledgement
I express my gratitude to my thesis supervisor, Paavo Nevalainen, my internship
supervisor, Tomi Westerlund, and my colleagues for allowing me to work with the
type of technology I am genuinely interested in. I value the opportunity to show-
case my skills and fully participate in the development of these systems for future
use. Working with the TIERS (Turku Intelligent Embedded and Robotic
Systems) team at the University of Turku provided hands-on experience that
significantly deepened my understanding of robotics, autonomous systems, and the
defense sector.
To note the use of artificial intelligence, Grammarly’s AI has been used to proof-
read and improve the thesis’s grammar.
i
Contents
0.1 Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i
1 Introduction 1
1.1 Thesis Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Thesis Research Questions . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Thesis Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.4 Thesis Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2 Background and Theory 6
2.1 Terminology of Maritime Drones (USVs, UUVs, ROVs, etc.) . . . . . 6
2.2 History of Unmanned Naval Systems . . . . . . . . . . . . . . . . . . 9
2.2.1 USVs - Unmanned Surface Vehicles . . . . . . . . . . . . . . . 10
2.2.2 UUVs - Unmanned Underwater Vehicles . . . . . . . . . . . . 15
2.3 Current Applications Sectors . . . . . . . . . . . . . . . . . . . . . . . 20
2.4 Maritime Drone Market . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.5 Technical Foundations . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.6 Maritime Drone Challenges . . . . . . . . . . . . . . . . . . . . . . . 26
3 Evaluating and Developing Maritime Drones 29
3.1 Analyzing Efficiency Improvements . . . . . . . . . . . . . . . . . . . 29
3.2 Key Performance Metrics . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.3 Tools, Models, and Frameworks Used in Development . . . . . . . . . 32
ii
4 Improving Military Maritime Drones 35
4.1 Energy Optimization and Advancements . . . . . . . . . . . . . . . . 35
4.2 Hydrodynamics and Design . . . . . . . . . . . . . . . . . . . . . . . 36
4.3 AI and Hardware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
4.4 Stealth Features and Communications . . . . . . . . . . . . . . . . . 43
4.4.1 Stealth Technologies . . . . . . . . . . . . . . . . . . . . . . . 44
4.4.2 Communication Security and Innovation . . . . . . . . . . . . 47
5 Maritime Drones in Modern Warfare 50
5.1 The Black Sea: Russo-Ukraine War . . . . . . . . . . . . . . . . . . . 50
5.2 The Gray Zone: The Red Sea, Persian Gulf, and the Israel–Gaza
Conflict . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
5.3 South and East China Sea: The Brewing Storm Between The U.S,
China, and Taiwan . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
6 Results 61
6.1 Strategic Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
6.2 Tactical Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
6.3 Emerging Technologies and Improvements . . . . . . . . . . . . . . . 64
6.4 Ethical and Environmental Concerns . . . . . . . . . . . . . . . . . . 66
6.5 Policy and Regulatory Considerations . . . . . . . . . . . . . . . . . . 68
6.5.1 European Union . . . . . . . . . . . . . . . . . . . . . . . . . . 69
6.5.2 National-Level Regulations . . . . . . . . . . . . . . . . . . . . 70
7 Conclusion 73
7.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
7.2 Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
References 76
iii
List of Figures
2.1 Unmanned surface vehicle classes defined by NAVSEA. [11] . . . . . . 8
2.2 Unmanned underwater vehicle classes defined by NAVSEA. [11] . . . 9
2.3 The first USV, Fernlenkboot, developed by the German Imperial
Navy. [5] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.4 The Protector USV on display at the National Museum of Singapore
in February 2014. [14] . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.5 The U.S. Navy’s Sea Hunter USV, developed by Vigor Industries. [15] 13
2.6 Ukraine’s first USV, the Mykola. [18] . . . . . . . . . . . . . . . . . . 14
2.7 Marlin, the first electronic warfare maritime drone. [16] . . . . . . . . 14
2.8 The first UUV, SPURV (1957), developed by the University of Wash-
ington. [21] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.9 NMRS system developed by the U.S. Navy. [22] . . . . . . . . . . . . 17
2.10 The REMUS 100 UUV, used by the Finnish Navy, on display in
Turku, 2016. [24] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.11 The Toloka-TLK150 UUV, developed by Brave1. [18] . . . . . . . . . 19
2.12 Copperhead UUVs, developed by Anduril Industries. [26] . . . . . . . 20
2.13 Projected growth of the global unmanned marine vehicles market. [27] 22
2.14 Key end-use segments in the unmanned marine vehicles market. [27] . 22
2.15 Navigation pipelines examples from the Rolls-Royce AAWA report.
[35] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
iv
4.1 Side profile of the Wave Rider drone. [40] . . . . . . . . . . . . . . . . 37
4.2 The Manta Ray drone as seen in Google Earth satellite imagery. [41] 38
4.3 The Manta Ray during testing in 2024. [25] . . . . . . . . . . . . . . 39
4.4 The world’s first hybrid drone, Triton. [42] . . . . . . . . . . . . . . . 40
4.5 From left: Petteri Hemminki, Waltteri Soininen, and Lauri Vasankari
presenting the Meri-Turso maritime drone board at the MPKK. [43] . 41
4.6 The Meri-Turso maritime drone prototype presented at the SecD-day
by Cpt. Jan Joutsi. [44] . . . . . . . . . . . . . . . . . . . . . . . . . 42
4.7 BAE Systems’ Adaptiv camouflage system rendered on a warship. [46] 45
4.8 Taiwan’s first military maritime drone. [49] . . . . . . . . . . . . . . . 47
4.9 6G Maritime relay network concept. [52] . . . . . . . . . . . . . . . . 49
5.1 Known Russian maritime drones used in the Black-Sea. [56] . . . . . 51
5.2 Known Ukrainian maritime drones used in the Black-Sea. [56] . . . . 52
5.3 Damage inflicted to the Kerch Bridge by a Ukrainian maritime drone.
[57] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
5.4 Ukrainian maritime drone during an attack on Olenegorsky Gornyak.
[55] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
5.5 Drone attacks in Ukraine during 2022-2023. [37] . . . . . . . . . . . . 54
5.6 Known maritime drones used by the Houthis. [60] . . . . . . . . . . . 56
5.7 The Blue Whale UUV, developed by Israeli Elta Systems. [61] . . . . 57
5.8 Successful Houthi attack on the Greek tanker MV Tutor, June 12,
2024 [6] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
5.9 Chinese military maritime activity during the August 2022 drill. [64] 59
6.1 Ukraine’s maritime drone, the Katran Venom. [65] . . . . . . . . . . 65
v
List of Tables
3.1 Key performance metrics for maritime drones. . . . . . . . . . . . . . 31
3.2 Comparison of simulators for maritime robotics. . . . . . . . . . . . . 34
vi
1 Introduction
We have entered the beginning stages of the age of autonomous warfare, and no
end is in sight. The Russo-Ukrainian war is an excellent example of how these
weapons perform in real-life use cases in the modern era. The Russo-Ukrainian
war has been a pivotal testing field for such systems and their development. Large
private robotics companies are deploying their products to Ukraine for experiments
on actual humans and enemy systems. Autonomous weapon systems are gathering
large amounts of data to improve their capabilities for potential future conflicts. We
have already seen the first fully unmanned multi-robot attack on Ukrainian soil and
many unmanned drones fighting against each other [1]. Ukraine’s Minister of Digital
Transformation, Mykhailo Fedorov, announced on May 29, 2025, that Ukraine paved
the way for fully autonomous strikes by conducting the first-ever fully autonomous
aerial drone attack [2]. Autonomous warfare will become a reality.
The field of autonomous drones has been growing staggeringly in research and
development. The required technology behind these systems has been available for
a long time. However, these have only gained popularity recently, primarily driven
by advancements in artificial intelligence and sensor technology.
This thesis explores the field of maritime drones, which are categorized based on
their operating environments or capabilities. The two main categories are Un-
manned Surface Vehicles (USVs) and Unmanned Underwater Vehicles
(UUVs). Chapter 2 provides a more detailed discussion of these terms and other
1.1 THESIS MOTIVATION 2
categorizations.
Maritime drones have transformed modern naval warfare by providing capabili-
ties such as logistics, intelligence, surveillance, reconnaissance, mine detection, mine
laying, anti-submarine warfare, and even offensive operations.
The challenges these types of drones have generated for the field have exploded in
quantity. They must navigate complex and harsh marine environments, support long
operational endurance, and function with minimal human intervention. However,
technological constraints such as power limitations, communication challenges, and
security vulnerabilities can hinder their full potential.
We can see from history that these systems date back to World War I. However,
the systems back then were not autonomous; they were remote-controlled. The de-
velopment of advanced autonomous maritime drones gained increased popularity in
the late 20th century. [3], [4] The recent conflicts between Russia and Ukraine, as
well as those between the Houthis and coalition forces, demonstrate how these sys-
tems can significantly influence asymmetric warfare. [5], [6] Asymmetric warfare
means that one belligerent of the war is less equipped than the other. However,
this does not imply that they cannot be used in symmetric warfare where the
opposing forces are nearly equal in military power.
1.1 Thesis Motivation
Maritime drones play an important role in modern warfare, and their significance is
becoming clearer as we witness advancements in their development and their ability
to operate effectively in various scenarios, like in the cases of the Black Sea, the Red
Sea, and the East and South China Seas. A key point to understand is that maritime
drones are used in many fields, which means there are various use cases beyond mili-
tary applications, including research, security, transportation, and other commercial
applications. Their ability to operate in the harsh marine environment makes them
1.2 THESIS RESEARCH QUESTIONS 3
perfect for these use cases. Despite the recent advancements in the maritime drone
field, there are remaining challenges in areas such as navigation, endurance, sensor
integration, and real-time data processing, which limit their full potential. While a
lot of research has been conducted mainly on aerial and ground drones, these mar-
itime drones present unique technological and operational challenges due to their
complex underwater and surface conditions.
My interest in these types of systems comes from both academic and practi-
cal experience in this field. My bachelor’s thesis, Tekoälyavusteiset asejärjestelmät
nykyaikaisessa sodankäynnissä, or AI-assisted weapon systems in modern warfare
in English, was an overview of all types of Lethal Autonomous Weapon Sys-
tems (LAWS) in modern warfare [7]. In this thesis, I decided to focus exclusively
on maritime drones. Additionally, my role as a reserve officer and my interest in
Open-Source Intelligence (OSINT) have led me to research the operational
needs of maritime drones for the defense sector. Overall, all of these experiences
have shaped my motivation to explore innovative solutions that can bridge the ex-
isting technological gaps related to maritime drones.
1.2 Thesis Research Questions
This thesis will answer three research questions on maritime drones, which are:
1. What are the strategic and tactical implications of maritime drones
in modern warfare?
2. How can maritime drones be optimized and equipped better?
3. What are the ethical and environmental considerations surrounding
the deployment of maritime drones? And what kind of regulatory
measures have been set for maritime drones?
1.3 THESIS OBJECTIVES 4
These research questions are necessary for defining the scope of the thesis. These
questions will serve as a guide for both the theoretical and the practical exploration
of the topic. The primary goal of this thesis is to understand the implications
that maritime drones have on the nature of modern warfare. This thesis aims to
contribute to the development of maritime drone systems that are more effective,
efficient, and ethically responsible, with benefits for both military and civilian ap-
plications.
1.3 Thesis Objectives
The main objectives of the thesis are:
1. Investigate the underlying technologies, systems, and processes involved in
maritime drone design, development, and deployment to deepen understanding
of them.
2. Study the use cases and influence of maritime drones currently used across
several sectors, including scientific, industrial, commercial, defense, and war,
but keep the main focus on warfare.
3. Explore potential improvements and changes in the design, efficiency, and
operational standards of maritime drones. Propose recommendations that
could contribute to more sustainable, affordable, and scalable solutions.
4. Provide a foundation for future studies by giving an overview of the current sit-
uation, identifying current challenges, and recommending areas where further
research and innovation could advance the field of maritime drone technology.
1.4 THESIS STRUCTURE 5
1.4 Thesis Structure
In this thesis, maritime drones, including USVs and UUVs, will be examined. The
focus will be on how these technologies have reshaped modern warfare, illustrated
through real-world case studies.
In Chapter 2, the technical properties for developing maritime drones, a historical
overview of their origins, and the terminology surrounding them will be explained.
Furthermore, it introduces the current challenges of maritime drones, the global
market, and their application fields.
In Chapter 3, approaches to evaluating maritime drones, considering features
like energy efficiency, lifespan, cost, and performance, will be explored. It will also
review tools, models, and frameworks used in developing maritime drones.
In Chapter 4, an overview of how maritime drones can be improved through en-
ergy optimization, hydrodynamics, artificial intelligence, stealth features, and com-
munication systems will be provided. This will be achieved mainly by comparing
different innovative maritime drones.
In Chapter 5, the use of maritime drones in modern warfare will be analyzed
by looking at ongoing conflicts worldwide. It will showcase the impact of them on
enemy assets, trade routes, and the rising tensions between nations.
In Chapter 6, the research questions set for the thesis will be answered. The
chapter will discuss the strategic and tactical impacts of maritime drones on mod-
ern warfare, efficiency improvements and combat enhancements, ethical and en-
vironmental issues, and regulatory measures such as the Geneva Convention, EU
regulations, and national laws.
In Chapter 7, the thesis will be concluded by summarizing the work done and
providing recommendations and final thoughts on future research, military applica-
tions, and the role of unmanned naval systems.
2 Background and Theory
This chapter will cover the fundamentals of maritime drones, exploring their origins,
various use cases, and the challenges associated with their development. Discuss the
environmental factors that contribute to these challenges. Moreover, the chapter will
examine the core technologies that enable the creation of naval maritime drones.
2.1 Terminology of Maritime Drones (USVs, UUVs,
ROVs, etc.)
Surrounding terminology of maritime drones can quickly become confusing due to
the various terms available. However, the main two categories are Unmanned
Surface Vehicles (USVs) and Unmanned Underwater Vehicles (UUVs).
The term autonomous can also be used instead of unmanned, leading to designa-
tions such as Autonomous Underwater Vehicles (AUVs) and Autonomous
Surface Vehicles (ASVs). Autonomous often implies that the vehicle possesses
self-decision capability, whereas unmanned may suggest that it is primarily operated
by a human but without an onboard crew. [3], [4]
Additionally, the term Remotely Operated Vehicle (ROV) is common in the
marine field. This term can be preceded by different prefixes, such as Remotely
Operated Underwater Vehicle (ROUV) or Remotely Operated Surface
Vehicle (ROSV). The prefix can also appear at the beginning, resulting in terms
2.1 TERMINOLOGY OF MARITIME DRONES (USVS, UUVS, ROVS, ETC.)7
like UROV or SROV. [4], [8]
NATO (North Atlantic Treaty Organization) has adopted the term Maritime
Unmanned Systems (MUS) to cover all maritime-related drones in a single cat-
egory. However, it is essential to note that MUS also includes aerial, ground, and
space systems related to maritime operations. [9] The EDA (European Defence
Agency) has used the term Unmanned Marine Systems (UMS), also including
air, sea, and space systems. [10]
Lastly, maritime drones can be categorized by both their mode of movement
and their size. This includes classifications such as gliders, bio-inspired robots, and
crawlers. Moreover, categorizing by size is widely used since drones can come in
various sizes. The Naval Sea Systems Command (NAVSEA) has categorized USVs
and UUVs into four classes by size, which can be seen in Figures 2.1 and 2.2 [11].
2.1 TERMINOLOGY OF MARITIME DRONES (USVS, UUVS, ROVS, ETC.)8
Figure 2.1: Unmanned surface vehicle classes defined by NAVSEA. [11]
2.2 HISTORY OF UNMANNED NAVAL SYSTEMS 9
Figure 2.2: Unmanned underwater vehicle classes defined by NAVSEA. [11]
Given the multitude of terms, it is more beneficial to focus on the capabilities and
specifications of each maritime drone rather than relying solely on the terminology.
Some maritime drones are classified as hybrid drones, capable of operating on the
surface, underwater, or even in the air. Additionally, maritime drones may have
specific terminology beyond what has already been introduced here.
2.2 History of Unmanned Naval Systems
To make it clear, only maritime drones that were or are remotely controlled or
autonomously operated will be covered here. The history of these systems date
2.2 HISTORY OF UNMANNED NAVAL SYSTEMS 10
back to World War I [5], which is quite impressive. Some have categorized human-
operated explosive suicide boats as maritime drones [5]. However, in this thesis,
they will not be covered. The same goes for ancient fireships that were lit and sent
towards a target [3]. This thesis will focus primarily on military maritime drones.
2.2.1 USVs - Unmanned Surface Vehicles
The first military maritime drones were remote-controlled, and already in 1898,
Nikola Tesla received a patent for a radio-controlled unmanned surface vehicle [12].
During World War I, remote-controlled boats carrying explosives were deployed in
the English Channel. The German Imperial Navy developed its version, called the
Fernlenkboot, as seen in Figure 2.3. At the same time, the British also experi-
mented with radio-controlled attacks on German naval ports. The Ferlenkboot was
not very successful during the war due to its limited range. Its 20-kilometer tether
restricted the range, and it had delay problems with the remote control. However,
despite these issues, it managed to strike two targets in 1917: the Nieuwpoort pier
and the HMS Erebus monitor of the Royal Navy of the United Kingdom of Great
Britain and Northern Ireland. Similar tactics continued in World War II. [5], [13]
2.2 HISTORY OF UNMANNED NAVAL SYSTEMS 11
Figure 2.3: The first USV, Fernlenkboot, developed by the German Imperial Navy.
[5]
After World War I, the development of military-based USVs almost halted. Sig-
nificant progress resumed in the early 2000s, notably in 2005 when Israel introduced
the Protector USV for the coalition forces operating in the Persian Gulf. The
Protector (see Figure 2.4) is built on a motorboat platform and is equipped with a
lightweight machine gun. [14]
2.2 HISTORY OF UNMANNED NAVAL SYSTEMS 12
Figure 2.4: The Protector USV on display at the National Museum of Singapore in
February 2014. [14]
In 2016, a serious step was taken by the Defense Advanced Research Projects
Agency (DARPA) when they introduced the Sea Hunter (see Figure 2.5), an anti-
submarine USV prototype, showing significant advancement in autonomous mar-
itime technology. The Sea Hunter can operate for months and travel thousands of
kilometers without any human involvement, making it fully autonomous. [15]
2.2 HISTORY OF UNMANNED NAVAL SYSTEMS 13
Figure 2.5: The U.S. Navy’s Sea Hunter USV, developed by Vigor Industries. [15]
In the 2020s, a lot of USV development from various parties started rapidly.
In 2020, Turkey developed their first armed USV prototype, the ULAQ [3], and
later in 2022, the world’s first-ever electronic warfare maritime drone, the Marlin
(see Figure 2.7) [16]. Additionally, in 2022, Ukraine developed its first kamikaze
maritime drones by introducing their first-generation USV, the Mykola (see Figure
2.6). In 2023, Russia developed its first kamikaze maritime drone, the Oduvanchik,
or Dandelion in English, which has similar capabilities to the Ukrainian SeaBaby
USV [17]. Ukraine has been developing kamikaze drones since the start of the larger-
scale attack from Russia in 2022. Russia has used maritime drones far less frequently
than Ukraine, despite having developed several of its own, since it is using more of
its traditional naval vessels.
2.2 HISTORY OF UNMANNED NAVAL SYSTEMS 14
Figure 2.6: Ukraine’s first USV, the Mykola. [18]
Figure 2.7: Marlin, the first electronic warfare maritime drone. [16]
Currently, there are hundreds of USVs developed by governments, militaries,
private companies, hobbyists, and researchers for various applications.
2.2 HISTORY OF UNMANNED NAVAL SYSTEMS 15
2.2.2 UUVs - Unmanned Underwater Vehicles
The development of military unmanned underwater vehicles started later than USVs,
with the Self-Propelled Underwater Research Vehicle (SPURV) by the Uni-
versity of Washington in 1957 to research the arctic waters. This drone could operate
for four hours and was like a small-scale submarine. [4], [19] Figure 2.8 shows the
SPURV. After the SPURV, similar projects continued to appear from various in-
stitutes. The Autonomous and Remote-Controlled Submarine (ARCS) by
ISE Ltd. was developed together with the Canadian Department of Defence and
Hydrographic Service in 1983. [20] Despite its name, it was only remote-controlled,
but it was able to operate for 35 hours with real-time communications, enabling
longer operations.
2.2 HISTORY OF UNMANNED NAVAL SYSTEMS 16
Figure 2.8: The first UUV, SPURV (1957), developed by the University of Wash-
ington. [21]
In the 1990s, the UUVs became more autonomous and capable after the Russian
Institute of Marine Technology started the development of the Solar Autonomous
Underwater Vehicle (SAUV I and SAUV II). The solar panels made it possi-
ble to have longer operations and equip them with more technology and payloads.
In 1995, the first glider UUVs appeared. Gliders were capable of operating for sev-
eral months since battery life was improved, and they used the ocean waves as a
way of moving and collecting energy back. [4] These two were primarily research-
focused, but they still considered military applications, as extended operational time
remained a key requirement. More on the military side, the U.S. Navy developed
2.2 HISTORY OF UNMANNED NAVAL SYSTEMS 17
the Near-Term Mine Reconnaissance System (NMRS) (see Figure 2.9), a
two-vehicle system launched from a submarine’s torpedo tube in 1996.
Figure 2.9: NMRS system developed by the U.S. Navy. [22]
In the 2000s, the first use cases in conflict were seen. A UUV was first utilized
in combat in 2003 during Operation Iraqi Freedom. The United States employed a
REMUS UUV to assist in the removal of sea mines from the region near the Umm
Qasr port. [23] For example, the Finnish Navy operates one version of this type of
UUV, the REMUS 100 UUV, as shown in Figure 2.10.
2.2 HISTORY OF UNMANNED NAVAL SYSTEMS 18
Figure 2.10: The REMUS 100 UUV, used by the Finnish Navy, on display in Turku,
2016. [24]
The 2020s is the most significant time for the development of UUVs and the first
broad use in modern warfare. In 2022, the successful attacks with USVs prompted
Ukraine to also develop UUVs. In 2023, the Ukrainian government-led agency
Brave1 developed three different-sized UUVs, called Toloka-TLK150, Toloka-
TLK400, and Toloka-TLK1000. The smallest model has a range of 100 kilo-
meters, while the largest can exceed 1,900 kilometers. In Figure 2.11, the smallest
Toloka-TLK150 is shown. [4]
2.2 HISTORY OF UNMANNED NAVAL SYSTEMS 19
Figure 2.11: The Toloka-TLK150 UUV, developed by Brave1. [18]
The industry in the United States followed after with their modern UUVs. First
with the Manta Ray by Northrop Grumman in 2024 and the Copperhead by
Anduril Industries in 2025. The manta ray is a bio-inspired extra-large maritime
drone, and the Copperhead family consists of two UUVs, the Copperhead-100 and
Copperhead-500, with the key difference being size, which introduces more room for
additional payloads and longer operational time. Copperheads can be launched from
Anduril’s Dive-LD and Dive-XL UUVs; both can carry multiple Copperheads.
Opening the possibility of a fully connected autonomous fleet. Anduril is already
collaborating with military partners on developing even more capable systems, such
as the Ghost Shark XL-AUV, which is expected to be operational in June 2025.
[25], [26]
2.3 CURRENT APPLICATIONS SECTORS 20
Figure 2.12: Copperhead UUVs, developed by Anduril Industries. [26]
2.3 Current Applications Sectors
Today, the maritime drone sector spans various sectors, and the capabilities of such
systems can differ. These drones play an increasingly important role in nearly all
aspects of the marine environment. Maritime drones are mostly employed currently
in [3], [4]:
• Military: Intelligence, surveillance, reconnaissance, mine detection and neu-
tralization, electronic warfare, logistics support, and force protection.
• Scientific: Oceanographic research, climate monitoring, and marine biodiver-
sity assessment.
2.4 MARITIME DRONE MARKET 21
• Industrial: Inspection of offshore infrastructure, monitoring of undersea ca-
bles, and detailed seabed mapping.
• Transportation: Autonomous ferries and logistics platforms.
The combination of autonomy, endurance, and adaptability makes them espe-
cially valuable in environments that are dangerous or inaccessible to human opera-
tors. Larger vessels can also use this type of autonomy. However, there is debate
about whether an autonomous ferry or cruise ship can be considered a drone; there-
fore, I will not specifically count them in this thesis as autonomous maritime drones.
2.4 Maritime Drone Market
The future of the maritime drone market, according to a recent report by Global
Market Insights, indicates that the market was valued at USD 4.8 billion in 2024
and is projected to grow at a compound annual growth rate (CAGR) of over
6.8% from 2025 to 2034 [27]. The growth of the maritime market is shown in Figure
2.13.
The market’s expansion is a result of a combination of technological advance-
ments, increasing defense budgets, and rising demand for ocean data collection and
offshore oil & gas exploration. Autonomous systems offer a low-cost and safer al-
ternative to traditional manned operations in harsh or inaccessible marine envi-
ronments. Furthermore, they provide a chart on how the market value is shared
according to the end application. From Figure 2.14, it is clear that the defense sec-
tor dominates the maritime drone market by over half, with a percentage of 56.8%.
[27]
2.4 MARITIME DRONE MARKET 22
Figure 2.13: Projected growth of the global unmanned marine vehicles market. [27]
Figure 2.14: Key end-use segments in the unmanned marine vehicles market. [27]
2.5 TECHNICAL FOUNDATIONS 23
These graphs illustrate the potential of maritime drones globally, not only in
military applications, though that remains the largest sector by far.
2.5 Technical Foundations
There are several core technologies used in maritime drones that are typical to any
robotics field: a mix of mechanical design, electronics, and software engineering.
The primary difference that lies in the maritime field is adapting these traditional
technologies to operate effectively in challenging and unpredictable marine environ-
ments. Key technologies that are used in maritime drones:
• Sensors and Data Acquisition:
Maritime drones rely on a variety of sensors, such as sonar, lidar, radar, GNSS,
and optical or thermal cameras. All of these are used to perceive and interpret
their environment. Sensors are foundational to real-time navigation, obsta-
cle avoidance, and mission-critical data collection. Sensors are integrated for
sensor fusion for more accurate and fail-proof systems. [28], [29]
• Communication Systems:
Effective communication is a critical component in the deployment of maritime
drones. In maritime drones, typical radio frequency (RF) communication is
used. While RF communication is effective at or above the water surface, it
becomes unreliable underwater. Therefore, underwater vehicles often depend
on acoustic communication. [28]
• Autonomous Navigation and AI:
Autonomous maritime drones integrate machine learning algorithms and AI-
driven control architectures to support decision-making, path planning, and
environment interpretation. These systems process sensor data using tech-
niques such as Simultaneous Localization and Mapping (SLAM), probabilistic
2.5 TECHNICAL FOUNDATIONS 24
reasoning, and reinforcement learning. Computer vision plays a vital role, en-
abling tasks like object recognition, horizon stabilization, and visual tracking
in dynamic marine settings. [30]
• Control Systems:
To ensure dynamic stability and accurate maneuverability, maritime drones
utilize control systems grounded in feedback principles. One of the most
widely used approaches is the Proportional–Integral–Derivative (PID) con-
troller, which continuously corrects system behavior based on deviations from
desired states. These controllers are mathematically robust and computation-
ally efficient, making them suitable for stabilizing depth, heading, and velocity
in turbulent marine environments. [29]
• Propulsion and Actuation:
Propulsion systems in maritime drones commonly include electric or gas-
powered motors driving propellers or thrusters. Often supplemented by control
surfaces such as rudders, fins, or vectored thrust mechanisms to allow precise
maneuvering. For underwater drones, propulsion systems are optimized for
minimal noise and efficient energy use to support stealth and long-duration
missions. [29]
• Power Systems:
Energy management is vital for prolonged, autonomous operation. Maritime
drones typically use rechargeable lithium-based batteries due to their high
energy density and stability. Some systems also incorporate renewable energy
sources like solar panels or energy harvesting from wave motion to enhance
endurance and reduce logistical dependence. [29] Efficient power distribution
and low-power computing hardware further improve operational longevity.
• Materials and Structural Design:
2.5 TECHNICAL FOUNDATIONS 25
The structural integrity of maritime drones is maintained through the use
of specialized materials such as marine-grade stainless steel, fiberglass com-
posites, and high-performance polymers. These materials offer resistance to
corrosion and biofouling while maintaining structural strength against water
pressure. Design considerations also focus on hydrodynamics, pressure toler-
ance, and modularity to facilitate maintenance and upgrades. [31]
• Software and Simulation:
Software ecosystems for maritime robotics often adapt general-purpose frame-
works, such as the Robot Operating System 2 (ROS2) or Mission Oriented
Operating Suite - Interval Programming (MOOS-IvP) for marine applications.
These platforms support mission scripting, autonomy modules, sensor integra-
tion, and real-time diagnostics. Simulation environments are widely used to
test navigation algorithms, assess control systems, and validate mission sce-
narios within virtual models of marine settings, helping to reduce risks before
real-world deployment. [32], [33]
• Cybersecurity:
As maritime drones become more connected, especially in military or commer-
cial contexts, cybersecurity is increasingly critical. Potential threats include
unauthorized access, signal spoofing, and data interception. Secure communi-
cation protocols, hardware encryption, and intrusion detection systems must
be integrated to ensure mission integrity and asset protection. [34]
The Rolls-Royce-led program, the Advanced Autonomous Waterborne Applica-
tions (AAWA) Initiative 2015-2017, has captured the similarity between ground-
based autonomous vehicles and maritime vessels quite well in their research on the
development of maritime autonomous vehicles. Looking at Figure 2.15, we can see
the two pipeline examples of an autonomous car and a maritime vessel. The two are
2.6 MARITIME DRONE CHALLENGES 26
similar, with mainly a difference in some sensors and the environmental differences
requiring different types of maps and charts. [35]
Figure 2.15: Navigation pipelines examples from the Rolls-Royce AAWA report.
[35]
The technology for maritime drones is already available. However, further de-
velopment and innovation are still necessary.
2.6 Maritime Drone Challenges
Maritime drones come with challenges. The harsh marine environment can change
rapidly, and it introduces critical aspects that need to be taken into account when
2.6 MARITIME DRONE CHALLENGES 27
developing maritime drones. To list common challenges:
Navigation and communication are some of the main challenges maritime drones
present. Most maritime drones depend on GPS navigation, which can be unreliable
in various situations. This is particularly true for underwater vehicles that cannot
effectively use GPS. As a result, these vehicles often need to rely on acoustic signals,
which can also be susceptible to interference. Maritime drones rely on network
connections to stay linked with their operators, process data via cloud computing,
or communicate and coordinate with other drones. Without stable connectivity,
tasks like remote control, real-time monitoring, and collaborative missions become
challenging to manage.
Power management and retrieving maritime drones is the next challenge. Mar-
itime drones operate far from shore, making it extremely difficult to recharge, per-
form maintenance, or retrieve them. For the maritime drones that are designed to
be expendable, like kamikaze types, retrieval is not an issue. Power management is
an issue, especially with naval drones that need to perform long-term operations or
have a lot of power consumption.
Temperature change and sensor accuracy are significant challenges for maritime
drones operating in both polar and equatorial waters, as they encounter conditions
ranging from freezing temperatures to extreme heat. Heat change, sediments, and
marine debris can affect the sensor’s performance. Additionally, high-pressure envi-
ronments are a challenge, as the intense pressure can crush the vehicle. [28], [29]
Military applications pose additional challenges for maritime drones. The mili-
tary forces of the enemy will often have similar maritime drones and counter systems
for them. Stealth and detection are critical factors to consider, as the drones must
operate without being detected while effectively carrying out the tasks assigned to
them. To achieve this, developers can minimize sensor output and prioritize the
use of more passive components to reduce electromagnetic signatures. Furthermore,
2.6 MARITIME DRONE CHALLENGES 28
materials that conceal temperature signatures or reduce radar detection are also im-
portant considerations. Cyberattacks or hijacking attempts by adversaries seeking
to gain control or sabotage the system should also be considered, not just in military
contexts. [34], [36]
The development process of maritime drones can be expensive. It follows a
standard software engineering approach while also emphasizing rapid innovation.
This means the product may undergo multiple prototype versions and several testing
phases to ensure everything operates seamlessly. After the product has proven to
work, in most cases, the development process will continue with improvements and
software updates. Even small-scale maritime drones can still exceed hundreds of
thousands in unit costs, not taking the development cost into account, like some of
Ukraine’s kamikaze drones [37].
Lastly, there are challenges regarding the application of regulatory measures to
maritime drones, as many regulations do not specifically mention them. Some regu-
lations may apply, but their enforcement largely depends on each nation’s willingness
to adhere to them. [38]
3 Evaluating and Developing
Maritime Drones
This chapter will cover the core metrics to evaluate the performance of maritime
drones and popular models, tools, and frameworks that assist in the development
process of maritime drones.
3.1 Analyzing Efficiency Improvements
When evaluating maritime drones, especially for military operations, certain main
features are sought after. Efficiency in maritime drones can refer to a range of
factors, including energy consumption, mission duration, navigation accuracy, data
throughput, and responsiveness to environmental conditions. When analyzing po-
tential improvements, the evaluation should be driven by the drone’s intended mis-
sion profile, whether it is a kamikaze drone, a USV or UUV, or some stationary
platform.
For military operations in particular, efficiency is closely tied to the following
aspects:
• Stealth: Acoustic signature, radar visibility, and heat emissions.
• Endurance: Battery or fuel efficiency for long-range deployment.
3.2 KEY PERFORMANCE METRICS 30
• Responsiveness: Ability to react and reroute based on dynamic threats or
changes in mission parameters.
• Modularity: Payload adaptability for missions such as surveillance,
reconnaissance, attack, and mine operations.
Analyzing the efficiency involves simulating and testing these parameters under
realistic environmental and operational scenarios. Additionally, specialized maritime
simulators, or wargaming, can be used.
3.2 Key Performance Metrics
To analyze maritime drone performance, I have gathered a Table 3.1 with some of
the main performance metrics that are looked after when analyzing maritime drone
capabilities and efficiency.
3.2 KEY PERFORMANCE METRICS 31
Feature Category Description
Hull design Stable and hydrodynamic shape, lightweight
and corrosion-resistant materials
Propulsion system Electric, fuel, or hybrid systems with efficient
propellers or thrusters
Navigation GPS/INS systems, sonar/lidar for obstacle
detection, autonomous path planning
Communication Satellite, RF, or acoustic links for control and
data transmission
Sensors and payload Cameras, sonar, radar, lidar; modular pay-
load configurations
Autonomy AI for mission planning, obstacle avoidance,
and adaptive behavior
Power management Long battery life, solar or wave energy, and
smart power allocation
Deployment Launch/recovery from ship or shore, modular
and portable design
Safety Emergency protocols, backup systems, mar-
itime safety compliance
Cost Low-cost is essential for maritime drones
Table 3.1: Key performance metrics for maritime drones.
3.3 TOOLS, MODELS, AND FRAMEWORKS USED IN DEVELOPMENT 32
These performance metrics are perfect for assessing design trade-offs and for
benchmarking different maritime drones from various parties.
3.3 Tools, Models, and Frameworks Used in Devel-
opment
Developing maritime drones requires several tools. Generally, the process follows a
standard software engineering process from the development perspective.
Frameworks are the starting point of developing a maritime drone. Frameworks
that are extremely popular currently for developing robots include Robotic Operat-
ing System 2 (ROS2), which is currently the biggest and most popular open-source
robotics framework. ROS2 is amazingly straightforward to use, and they have exten-
sive documentation. [32] The basic architecture behind ROS2 is that nodes transfer
data between each other. Nodes are either publishers or subscribers, and they can
be extended to servers for more complex functionality. Another excellent framework
especially designed for maritime drones is Mission Oriented Operating Suite-Interval
Programming (MOOS-IvP). MOOS-IvP is a comprehensive, modular, open-source,
C++-based autonomy framework developed and maintained by the MIT Marine
Robotics Laboratory [33]. MOOS-IvP’s highly modular architecture makes it an
excellent choice, but it requires far more knowledge than ROS2. Moreover, one
choice is to program everything without any framework. However, doing it this
way will require much more development, which increases the cost and time used
immediately.
Simulators are useful for testing robots before they operate in the real world.
There are universal simulators, and some simulators have been specially developed
for the maritime robotics community. The main simulators used in robotics are
Gazebo, Webots, RViz, and Unity, which have compatibility with ROS2. Addition-
3.3 TOOLS, MODELS, AND FRAMEWORKS USED IN DEVELOPMENT 33
ally, smaller open-source simulators exist, such as Stonefish and HoloOcean, which
are more physics-based and try to capture the actual hydrodynamical physics. Each
simulator has its pros and cons. Simulators accelerate the development process by
eliminating the need for actual hardware, allowing a single developer to test various
scenarios and environments. Simulators can emulate various sensors, making them
excellent tools to try different configurations. During my research, I explored various
simulators and summarized my findings in Table 3.2.
When simulating robots, a robot model is necessary. In ROS2, for instance, the
Unified Robot Description Format (URDF) is used to create 3D models of
robots [32]. Robotics engineers mostly use professional computer-aided design
(CAD) programs for designing mechanical and electrical parts. The most popular
mechanical CAD programs are SolidWorks, Catia, and Autodesk. Additional plugins
are available to convert these designs to URDF or include them in ROS2 simulators.
Some simulators can use various CAD file extensions.
Some robots might need custom PCBs and electrical design. Popular electrical
CAD programs are KiCad, Altium Designer, Eagle, and EasyEDA. These are by far
the most prominent and easiest to learn for beginners. These tools allow designers
to create custom PCBs, design electrical schematics, and simulate them to verify
functionality before building the robot.
3.3 TOOLS, MODELS, AND FRAMEWORKS USED IN DEVELOPMENT 34
S
im
u
la
to
r
A
d
va
nt
ag
es
L
im
it
at
io
n
s
G
az
eb
o
–
N
at
iv
e
in
te
gr
at
io
n
w
it
h
R
O
S2
–
R
ob
us
t
se
ns
or
si
m
ul
at
io
n
an
d
m
od
ul
ar
ar
ch
it
ec
tu
re
–
E
xt
en
si
ve
pl
ug
in
su
pp
or
t
–
C
om
pu
ta
ti
on
al
ly
in
te
ns
iv
e
–
R
eq
ui
re
s
cu
st
om
ex
te
ns
io
ns
fo
r
re
al
is
ti
c
m
ar
in
e
ph
ys
ic
s
H
ol
oO
ce
an
–
B
ui
lt
es
pe
ci
al
ly
fo
r
un
de
rw
at
er
en
vi
ro
nm
en
ts
–
In
cl
ud
es
m
os
t
m
ar
it
im
e
se
ns
or
s
–
M
ul
ti
-a
ge
nt
se
tu
ps
–
C
om
pl
ex
se
tu
p
an
d
de
pe
nd
en
cy
m
an
ag
em
en
t
–
Li
m
it
ed
do
cu
m
en
ta
ti
on
U
n
it
y
+
M
A
R
U
S
–
H
ig
h-
qu
al
ity
gr
ap
hi
cs
–
C
us
to
m
iz
ab
le
3D
en
vi
ro
nm
en
ts
–
R
ea
l-t
im
e
si
m
ul
at
io
n
ca
pa
bi
lit
ie
s
–
Lo
w
ph
ys
ic
al
re
al
is
m
–
R
eq
ui
re
s
sc
ri
pt
in
g
fo
r
ac
cu
ra
te
m
ar
in
e
ph
ys
ic
s
W
eb
ot
s
–
Li
gh
tw
ei
gh
t
an
d
us
er
-fr
ie
nd
ly
–
O
pe
n-
so
ur
ce
an
d
w
el
l-d
oc
um
en
te
d
–
Si
m
pl
ifi
ed
hy
dr
od
yn
am
ic
m
od
el
in
g
–
Li
m
it
ed
gr
ap
hi
ca
la
nd
se
ns
or
re
al
is
m
S
to
n
efi
sh
–
D
es
ig
ne
d
sp
ec
ifi
ca
lly
fo
r
m
ar
in
e
ro
bo
ti
cs
–
M
od
ul
ar
an
d
in
te
gr
at
es
w
it
h
R
O
S2
–
St
ill
un
de
r
ac
ti
ve
de
ve
lo
pm
en
t
–
Sm
al
le
r
ec
os
ys
te
m
co
m
pa
re
d
to
le
ga
cy
pl
at
fo
rm
s
Ta
bl
e
3.
2:
C
om
pa
ri
so
n
of
si
m
ul
at
or
s
fo
r
m
ar
it
im
e
ro
bo
ti
cs
.
4 Improving Military Maritime
Drones
This chapter will explore a range of practical advancements that have been imple-
mented to enhance the operational efficiency of maritime drones. These improve-
ments will be categorized into seven distinct areas, each addressing a key aspect of
performance optimization, technological innovation, or environmental adaptability.
4.1 Energy Optimization and Advancements
Maritime drones mostly rely on traditional energy sources for power. The demand
for longer operation times and the integration of additional devices and weapons
necessitates an increase in power consumption. Therefore, it is important to consider
each individual component since they play a role in the scheme of everything. Well,
one could think that there is no way of optimizing energy efficiency, but that is not
true. Despite humanity having found various ways of generating power, there is still
a long road ahead for improvements to come. Some areas currently under heavy
innovation are battery technology, solar power, wave energy harvesting, hydrogen
cells, and thermal energy. All of these methods can be used on maritime drones to
enhance their energy efficiency.
The improvement does not need to be a physical feature. Energy optimization
can be achieved by developing more efficient software. AI tasks can be sent to
4.2 HYDRODYNAMICS AND DESIGN 36
the cloud, which is called task offloading, and path planning can be designed in
an energy-efficient way [28]. Different power management modes and scheduling
can be implemented. Sensors can be strategically disabled, or dynamic loading can
be utilized, meaning the system only consumes the necessary power. Additionally,
multi-robot systems can influence energy efficiency by sharing information to collab-
orate in optimizing paths, potentially leading to more effective energy distribution
[28].
On the hardware side, sensors can be designed for low power consumption to min-
imize energy usage; excellent examples of these are Field Programmable Gate Arrays
(FPGAs) and neuromorphic processors. Furthermore, actuators can be picked with
minimal power consumption, and energy can be collected back with energy harvest-
ing from solar, wind, and wave energy [29].
4.2 Hydrodynamics and Design
To improve the efficiency of maritime drones, the principles of hydrodynamics can
be applied. Hydrodynamics is an excellent way to affect the behavioral aspects of a
maritime drone with water. [39] Hydrodynamics can be used in clever ways, and I
will introduce four different maritime drones that have been specially designed with
hydrodynamics and efficiency in mind.
The Wave Glider (Figure 4.1), developed by Liquid Robotics, a subsidiary
of The Boeing Company, uses an innovative propulsion method that harnesses the
energy of ocean waves. It has two main sections. The top section stays on the
surface, and it can have various payloads. The bottom section is tethered to the top
section, and the bottom section is submerged eight meters underwater. The bottom
section is responsible for utilizing the wave energy to glide the drone to its target.
Liquid Robotics states that the Wave Glider can operate for several months. [40]
4.2 HYDRODYNAMICS AND DESIGN 37
Figure 4.1: Side profile of the Wave Rider drone. [40]
The Manta Ray UUV (Figure 4.3) by Northrop Grumman that was developed
in 2024 is shaped like a manta ray due to their efficient movement underwater. Manta
Ray is an excellent example of a biology-based maritime drone. While the Manta
Ray cannot flex its body like a real manta ray, it mimics the animal’s hydrodynamic
movement. The manta ray can operate autonomously for long periods, and it can be
customized to different sizes and operations, mainly to be used as a logistics drone.
[25]
4.2 HYDRODYNAMICS AND DESIGN 38
Figure 4.2: The Manta Ray drone as seen in Google Earth satellite imagery. [41]
4.2 HYDRODYNAMICS AND DESIGN 39
Figure 4.3: The Manta Ray during testing in 2024. [25]
The Triton (Figure 4.4) is one of the first hybrid maritime drones that can op-
erate both on the surface and underwater. It has been fully developed by OceanAero
in the United States. The drone has a foldable sail and is fully covered with solar
panels, enabling it to sail for up to three months. By folding the sail down, it
can operate for five days underwater. OceanAero themselves describe the Triton
as the world’s first Autonomous Underwater and Surface Vehicle (AUSV), and it
has also been referred to as the Sail Submarine. The main purpose of Triton for
military applications is to provide intelligence, surveillance, reconnaissance, mine
countermeasures, and anti-submarine warfare. [42]
4.2 HYDRODYNAMICS AND DESIGN 40
Figure 4.4: The world’s first hybrid drone, Triton. [42]
The Finnish National Defense University (MPKK), together with our
team from Turku University and Turku University of Applied Sciences, has
developed an efoil board-based maritime drone, the Meri-Turso. Figure 4.5 shows
the bottom section of the Meri-Turso drone [43]. From this view, it is clear that
the design of Meri-Turso supports efficient hydrodynamic performance. The design
is based on flying atop a submerged wing, which generates lift to keep the vehicle
above the water. This configuration minimizes the contact with water, thereby
significantly reducing drag. This design also helps to remove cavitation in the case
that Meri-Turso is always moving at the optimal depth. Traditional propellers
are often placed near the surface, or they spin too fast, which causes cavitation.
Cavitation can be problematic for military maritime drones, as the hissing and
bubble popping can be detected by enemy passive sensors. A noisy drone is more
4.2 HYDRODYNAMICS AND DESIGN 41
likely to be detected, tracked, or targeted. Meri-Turso’s low-rpm and high-torque
motor minimizes cavitation, providing a constant flow in water.
The main purpose of Meri-Turso is to provide remote surveillance. However, use
cases can vary from mission to mission, and the drone could even be equipped with
an explosive payload to function as a kamikaze drone. Additionally, it could operate
in swarms, making it almost impossible to counter. Meri-Turso uses intelligent
control systems to stabilize itself in various situations and can navigate with ease.
The first built prototype (see Figure 4.6) by our team of universities was presented
to the public at the SecD-day event in Helsinki Expo & Convention Center on 29-30
January 2025 [44].
Figure 4.5: From left: Petteri Hemminki, Waltteri Soininen, and Lauri Vasankari
presenting the Meri-Turso maritime drone board at the MPKK. [43]
4.3 AI AND HARDWARE 42
Figure 4.6: The Meri-Turso maritime drone prototype presented at the SecD-day
by Cpt. Jan Joutsi. [44]
4.3 AI and Hardware
Since the beginning of artificial intelligence (AI) in the 1950s, the field has progressed
dramatically due to advances in both computing platforms and algorithmic design.
Better computing platforms and algorithms have been pivotal for the research and
development of AI and autonomy in robotics. As a core element of naval drones,
artificial intelligence significantly improves system efficiency. In such systems, AI
is critical to achieving efficient navigation, localization, object detection, threat as-
sessment, and mission execution. Numerous ways to implement AI algorithms, each
with trade-offs in terms of speed, accuracy, and computational cost, are available
[30]. As these algorithms continue to evolve, they are increasingly well optimized
for performance and robustness in real-world conditions.
AI tasks demand a significant amount of processing power, especially for real-
4.4 STEALTH FEATURES AND COMMUNICATIONS 43
time inference tasks such as computer vision and sensor fusion, and therefore, opti-
mal hardware is needed. In the context of military maritime drones, these need to be
relatively cheap and capable of handling rough environments and possible jamming
from electronic warfare systems.
Single-board computers (SBCs) are a common choice in robotics due to their
compact size and integrated capabilities. Two widely used SBC platforms are the
NVIDIA Jetson series and the Raspberry Pi. The Jetson family shows better per-
formance in graphically intensive applications because it is designed for AI tasks. In
contrast, the Raspberry Pi is considered more a hobbyist-level SBC, but it has also
proven successful in various robotics projects. [45]
For defense applications, ruggedized embedded computing platforms designed es-
pecially for military environments are often preferable. These systems are typically
tested and rated for shock, vibration, temperature extremes, and electromagnetic
compatibility. In the case of kamikaze drones, the cost should be minimized. There-
fore, rugged computers might not be the right choice for that use case.
Ultimately, the choice of computational hardware in maritime drones must con-
sider at least three core aspects: computational requirements, environmental re-
silience, and cost-effectiveness. The future of AI-powered naval systems will depend
heavily on continued development in both software and hardware platforms.
4.4 Stealth Features and Communications
Stealth and secure communication technologies are critical features for modern mil-
itary operations. As unmanned systems become more prevalent in naval warfare,
ensuring their survivability, discretion, and secure command and control capabilities
is essential. Maritime drones, especially those intended for reconnaissance, surveil-
lance, or offensive operations, must be equipped with sophisticated stealth features
to avoid detection and advanced communication systems to function in contested or
4.4 STEALTH FEATURES AND COMMUNICATIONS 44
denied environments. I will introduce three points for each section.
4.4.1 Stealth Technologies
The marine environment again poses unique challenges for stealth due to factors
such as sonar, radar, infrared detection, and electromagnetic signature tracking.
In response, several advanced stealth techniques are being explored or are to be
implemented in maritime drones:
1. Adaptive Thermal Camouflage
Adaptive thermal camouflage would enable unmanned vessels to regulate or
disguise their heat signatures to evade infrared detection from enemy sen-
sors. Using materials or coatings that respond dynamically to environmental
temperatures would allow drones to blend into the sea surface’s thermal back-
ground. Adaptive thermal camouflage would be particularly useful in avoiding
thermal imaging sensors used by enemy aircraft, satellites, or coastal surveil-
lance systems. The ADAPTIV camouflage was introduced in 2011 by BAE
Systems AB in Sweden, and it has been tested on CV90 infantry fighting
vehicles with excellent results. Figure 4.7, provided by BAE Systems, illus-
trates how ADAPTIV technology could be applied to a warship, suggesting
its potential use on maritime drones as well.
4.4 STEALTH FEATURES AND COMMUNICATIONS 45
Figure 4.7: BAE Systems’ Adaptiv camouflage system rendered on a warship. [46]
2. Electromagnetic Signature Reduction
Naval vessels emit a range of electromagnetic signals. Electromagnetic signals
can be emitted from propulsion systems, onboard electronics, and communica-
tion equipment. To remain undetected, maritime systems should be engineered
with electromagnetic shielding and emission control protocols (EMCON) in
mind. [47] Features like these help reduce or mask emissions, making it harder
to trigger radar or electronic surveillance systems. Signaling can be configured
in timed bursts or limited windows to avoid continuous transmission. Addi-
tionally, to reduce the electromagnetic signature, more passive sensors that do
not emit signals can be used.
3. Radar cross-section (RCS) reduction
Radar cross section can be minimized in four ways [48]:
(a) Shaping
4.4 STEALTH FEATURES AND COMMUNICATIONS 46
(b) Radar-absorbing materials (RAM)
(c) Passive cancellation
(d) Active cancellation
Maritime drones may be constructed using composites or coatings that ab-
sorb rather than reflect radar waves. Radar-absorbing materials help reduce
the RCS, making unmanned vessels more difficult to track or identify using
conventional radar systems. These materials are already in use on stealth air-
craft and submarines and should be adapted for small surface and subsurface
drones. Another way of reducing RCS is to have low-profile or wave-skimming
hull designs. Some maritime drones employ small radar cross-section hulls or
semi-submersible profiles that allow them to stay below the horizon and radar
line of sight, further increasing stealth in secret operations. Passive and active
cancellation can be used to either echo the waves back or generate waves that
counter the radar’s waves. Passive and active cancellation could be used for
maritime drones. Shaping can be seen in many traditional naval vessels but is
not yet in broad use for maritime drones. However, for example, Taiwan’s first
maritime drone or the U.S. Navy’s Sea Hunter are excellent examples of such
shaping; see Figures 4.8 and 2.5. These RCS improvements usually also come
with the negative aspects of reduced payload, reduced range, added weight,
and possibly increased maintenance [48].
4.4 STEALTH FEATURES AND COMMUNICATIONS 47
Figure 4.8: Taiwan’s first military maritime drone. [49]
4.4.2 Communication Security and Innovation
Reliable and secure communications are essential for unmanned or autonomous
drones, especially in high-risk settings. However, traditional communication meth-
ods are vulnerable to jamming, interception, and spoofing. A few advanced tech-
nologies are being developed to mitigate these vulnerabilities:
1. Adaptive AI-Based Communication Protocols
Artificial intelligence could be used to manage drone communication systems
more flexibly and intelligently. AI can autonomously shift frequencies, detect
interference, and reroute communication paths to maintain the network as
operational [50]. This capability would allow maritime drones to maintain
command-and-control links even under an electronic attack. These systems
4.4 STEALTH FEATURES AND COMMUNICATIONS 48
can adapt in real-time to emerging threats and environmental conditions.
2. Mesh Networking Among Other Drones
Next-generation drones should be equipped with more mesh networking ca-
pabilities, allowing them to communicate with each other directly. Enabling
data relays over large distances, improves redundancy, and reduces reliance on
a single ground control station, which is especially valuable for coordinated
fleet operations. In 2023, one such 5G mesh networking system was tested
by NATO in the Dynamic Messenger test [51]. This type of mesh connection
can be seen in Figure 4.9, where especially the UUVs are connected with each
other.
3. 4G - 6G, Satellite, and Other Communication Integrations
Modern maritime drones often switch between traditional radio frequency com-
munications and satellite uplinks depending on range and environment. The
integration of multiple channels allows for redundancy and greater resilience
to signal loss or jamming. Both Ukraine and Russia have used the Starlink
terminals developed by SpaceX on their maritime drones [3]. Starlink works
by providing internet access through satellite communication. If satellite com-
munication is unavailable, the system will fall back to a Long-Term Evolution
(LTE) or, as it is known, 4G connection. Furthermore, since 5G and 6G net-
works have proved to work quite well in other fields, they should be utilized
more in maritime drones by having special maritime drones to relay the signal
or link buoys. Figure 4.9 illustrates a relay-based 6G network for maritime
drones.
4.4 STEALTH FEATURES AND COMMUNICATIONS 49
Figure 4.9: 6G Maritime relay network concept. [52]
5 Maritime Drones in Modern
Warfare
5.1 The Black Sea: Russo-Ukraine War
Ukraine has countered the Russian naval fleet with a range of maritime drones.
Ukraine does not have a well-equipped navy since Russian forces have captured
most of the Ukrainian naval bases during the battles from 2014 to 2025. Therefore,
Ukraine developed maritime drones to combat Russian vessels and even established
a specialized brigade for USVs in August 2023, the 385th Unmanned Surface Vehicles
Brigade [53]. Ukraine has already managed to destroy a third of the Russian naval
fleet with maritime drones and missiles, according to the Security Service of Ukraine
(SBU) [54].
Some larger-scale operations were so significant that the Russian naval fleet de-
cided to move their vessels back to their home bases. In 2023, Ukraine sank the
Russian landing ship Olenegorsky Gornyak by attacking it with several USVs [55].
In 2024, two notable maritime drone attacks were successfully carried out: the
downing of a Russian Mi-8 helicopter and an attack on the Kerch Bridge. The
damage inflicted by these maritime drones to the Kerch Bridge and the landing ship
Olenegorsky Gornyak are shown in Figures 5.3 and 5.4. Numerous other successful
operations have also targeted Russian oil platforms, military seaports, and various
5.1 THE BLACK SEA: RUSSO-UKRAINE WAR 51
strategic assets throughout the war. Moreover, on May 2, 2025, we saw yet more
downings of two Russian SU-30 fighter jets from Ukraine’s MAGURA-7 USV with
U.S.-made AIM-9 Sidewinder Missiles [54].
The use of maritime drones in Ukraine is, by far, the most extensively docu-
mented example of their deployment in a real-world asymmetric modern conflict.
The tactics Ukraine has used are marine drone swarms and explosive-laden USVs,
with the help of remote piloting via Starlink and similar satellite-based communi-
cation systems. All successful attacks from 2022 to 2023 are shown in Figure 5.5.
Although only five attacks were reported during this period, the figure illustrates
that maritime drones have been effective in Ukraine, primarily due to the strategic
importance of each operation.
Furthermore, the drones currently known to be used by Ukraine and Russia
can be seen in Figures 5.1 and 5.2. By looking at these figures, it is clear that the
Ukrainian armed forces have utilized more maritime drones and equipped them with
a greater variety of weapon systems than Russia.
Figure 5.1: Known Russian maritime drones used in the Black-Sea. [56]
5.1 THE BLACK SEA: RUSSO-UKRAINE WAR 52
Figure 5.2: Known Ukrainian maritime drones used in the Black-Sea. [56]
Figure 5.3: Damage inflicted to the Kerch Bridge by a Ukrainian maritime drone.
[57]
5.1 THE BLACK SEA: RUSSO-UKRAINE WAR 53
Figure 5.4: Ukrainian maritime drone during an attack on Olenegorsky Gornyak.
[55]
5.2 THE GRAY ZONE: THE RED SEA, PERSIAN GULF, AND THE
ISRAEL–GAZA CONFLICT 54
Figure 5.5: Drone attacks in Ukraine during 2022-2023. [37]
5.2 The Gray Zone: The Red Sea, Persian Gulf,
and the Israel–Gaza Conflict
In the context of escalating regional tensions following the Israel-Gaza conflict,
Yemen’s Houthi movement has intensified its maritime warfare strategy by also
incorporating maritime drones into its operational toolkit. The currently known
arsenal used by the Houthis is shown in Figure 5.6. A report by Reuters in 2024
details the group’s shift toward deploying explosive-laden maritime drones to target
commercial shipping transiting the southern Red Sea and Bab al-Mandab Strait, a
globally critical maritime corridor since 12-15% of worldwide trade goes through the
5.2 THE GRAY ZONE: THE RED SEA, PERSIAN GULF, AND THE
ISRAEL–GAZA CONFLICT 55
strait and the Red Sea.
Since November 2023, the Houthis have conducted numerous maritime drone
attacks, with at least forty-three vessels targeted and twenty-one confirmed as dam-
aged by direct strikes. However, only thirteen reported being attacked by a drone.
While their earlier operations relied on boarding actions and projectile weaponry,
the use of USVs marked a notable evolution in their asymmetric maritime tactics.
Operating in the confined waters of the Red Sea and exploiting the dense shipping
traffic in the Bab al-Mandab Strait, the drones are difficult to detect and intercept
and can be launched from concealed coastal or mobile platforms. The success of
such attacks has not only inflicted material damage, including the sinking of at least
one bulk carrier, but also severely disrupted international maritime logistics. Ma-
jor shipping companies have diverted their ships around the Cape of Good Hope,
adding substantial costs and delays to global trade. [58]
A large supporter of the Houthis is Iran, which uses them as a proxy. The
Houthis have also targeted the U.S. Navy multiple times, but none of the attacks
have been successful since the U.S. Navy has countered every attack with ease. Iran
has done its attacks more in the Persian Gulf area, also targeting international cargo
ships and oil tankers or even the U.S. Navy. On June 12, 2024, the Houthis attacked
the U.S. Navy and a Greek-owned MV-Tutor tanker. The U.S. Navy was able to
destroy six of the maritime drones launched by the Houthis, but one managed to
strike the Greek tanker, which can be seen in Figure 5.8.
Israel has sabotaged and surveilled Gaza’s Hezbollah maritime supply lines with
UUVs during the larger conflict. And likewise, Hamas and Hezbollah have received
maritime drones from Iran. These drones are constructed with materials designed to
evade radar detection and are capable of approaching Israeli coastal targets. Hamas
has also been developing maritime drones for potential attacks on Israeli maritime
assets. Israel has been able to counter the proxies quite well with its marine drone
5.2 THE GRAY ZONE: THE RED SEA, PERSIAN GULF, AND THE
ISRAEL–GAZA CONFLICT 56
fleet. One notable attack allegedly attributed to Hamas was the cutting of under-
water communication cables in the Red Sea, which disrupted 25% of the region’s
internet connectivity. However, no official confirmation has been issued regarding
who was responsible for the incident. On the other hand, Israel, with a much smaller
navy compared to Iran, has effectively repelled attacks and is developing its UUVs.
Blue Whale (see Figure 5.7) is probably the best-known drone they have devel-
oped for intelligence, submarine detection, and reconnaissance. Additionally, the
maritime drones that Iran produces are cheaper and faster to build than the ones
that Israel is building. [59]
Figure 5.6: Known maritime drones used by the Houthis. [60]
5.2 THE GRAY ZONE: THE RED SEA, PERSIAN GULF, AND THE
ISRAEL–GAZA CONFLICT 57
Figure 5.7: The Blue Whale UUV, developed by Israeli Elta Systems. [61]
Figure 5.8: Successful Houthi attack on the Greek tanker MV Tutor, June 12, 2024
[6]
5.3 SOUTH AND EAST CHINA SEA: THE BREWING STORM BETWEEN
THE U.S, CHINA, AND TAIWAN 58
5.3 South and East China Sea: The Brewing Storm
Between The U.S, China, and Taiwan
Tensions in the South and East China Seas have seen interceptions and shadowing
involving USVs and UUVs, but not full-on attacks yet. In this region, maritime
drones are already being prepared for possible future conflicts. Sitting at the cross-
roads of major sea lanes and near contested maritime zones, Taiwan would be a
central player in any U.S.-China conflict, especially given the United States’ heavy
reliance on Taiwan for advanced electronics and semiconductor production. Taiwan
is responsible for manufacturing at least 90% of the world’s computer chips [62].
China continues to expand its drone capabilities in the region, while Taiwan is
developing its unmanned systems for coastal defense. In the event of a crisis, these
sea areas would likely serve as a key operational zone, making maritime drones
vital tools for surveillance, denial, and early warning on all sides. The U.S. Navy
maintains a significant presence in the region, leading to encounters between Chinese
authorities and U.S. Navy maritime drones. In December 2016, tensions sparked
quickly when China decided to seize one of the U.S. Navy’s UUVs from international
waters [63].
The tensions have been rising steadily between the nations, and Chinese military
forces have been training for the occupation of Taiwan multiple times with various
military drills. Furthermore, the Chinese forces have tried to scare Taiwan by op-
erating maritime drones around the country. Figure 5.9 shows how the Chinese
military utilized maritime drones in their military drills in August 2022. From the
image, we can see that China launched the drones from the East China Sea and
used the Japanese islands as cover to initiate their simulated attack and operations
on the eastern side of Taiwan before returning via the same route. Once again,
violating Taiwan’s territorial waters. [64]
5.3 SOUTH AND EAST CHINA SEA: THE BREWING STORM BETWEEN
THE U.S, CHINA, AND TAIWAN 59
Figure 5.9: Chinese military maritime activity during the August 2022 drill. [64]
I included this case study because this region has the potential to become a new
area of maritime warfare. Some military experts warn that it could even serve as
a potential flashpoint for World War III. It is important to emphasize that there is
currently no active conflict in the area. However, the ongoing arms race between
nations is evident. Military developments are visible on both sides, and China’s
drills are increasingly hostile toward Taiwan. Furthermore, relations between China
5.3 SOUTH AND EAST CHINA SEA: THE BREWING STORM BETWEEN
THE U.S, CHINA, AND TAIWAN 60
and the United States remain tense.
6 Results
This chapter will address the research questions outlined in Chapter 1 by examining
emerging technologies for maritime drones and their strategic and tactical implica-
tions. Additionally, it will provide a brief overview of the ethical and environmental
issues associated with maritime drones, as well as the regulatory measures that have
been established for their use.
6.1 Strategic Implications
By looking at the use of maritime drones in modern warfare settings, I have divided
the strategic and tactical implications into four smaller categories. As seen from
the previously presented drones and use cases in modern warfare in Chapter 5, it
can be concluded that maritime drones play a significant role in both strategic and
tactical implications. The surface and underwater domains have opened two new
fronts in warfare for autonomous drones. Furthermore, this and the following section
should be considered as an answer to the first research question: 1. What are the
strategic and tactical implications of maritime drones in modern warfare?
1. Shift in Naval Power Dynamics
Maritime drones are enabling the redistribution of influence. Unlike traditional
warships, which require substantial investment, maintenance, and personnel,
maritime drones are relatively inexpensive and accessible. Allowing smaller
6.1 STRATEGIC IMPLICATIONS 62
states, and even non-state actors, to bear maritime drones. Resulting in tra-
ditional naval power being more vulnerable to threats posed by these drones.
This shift is evident in recent conflicts, where smaller forces have effectively
employed maritime drones to challenge naval powers.
2. Deterrence and Escalation
The expansion of the maritime drone field complicates the strategic calculus
around deterrence and escalation. These systems are well suited for so-called
gray zone operations. These are actions that fall below the threshold of an
open conflict but can still alter the strategic balance, such as seen in the Red
Sea and East and South China Seas. Since these drones can operate without
direct human risk and with plausible deniability, they offer states a tool to
probe, harass, and even damage adversaries without necessarily provoking a
full-scale military response. This blurring of boundaries raises new discussions
about intent, attribution, and proportionality in maritime conflict.
3. Strategic Surveillance and Sea Denial
One of the most significant strategic implications of maritime drones is their
ability to conduct persistent intelligence, surveillance, and reconnaissance op-
erations. UUVs, in particular, can operate for long periods while remaining
undetectable. This capability improves maritime domain awareness, especially
in contested regions or near strategic chokepoints. By integrating maritime
drones into anti-access/area denial operations, states can bolster their ability
to deter or obstruct hostile naval activity within critical sea lanes and their
territorial waters.
4. Arms Race and Naval Modernization
The development and use of maritime drones are fueling a new naval arms race.
Major military nations, including the United States, China, and Russia, are
6.2 TACTICAL IMPLICATIONS 63
investing billions in autonomous naval systems. This advancement applies not
only to hardware but also to the integration of artificial intelligence, swarming
capabilities, and advanced underwater navigation and communication systems.
As this arms race intensifies, there is a risk of destabilization, especially as
these technologies become more autonomous and less predictable black boxes.
6.2 Tactical Implications
Once again, I have divided the tactical implications into four categories, which I
consider to be the most significant:
1. Force Multiplication
At the tactical level, maritime drones offer an excellent force multiplier ef-
fect. Multiple of drones can be deployed simultaneously to perform coordi-
nated maneuvers, reconnaissance, or attacks. In swarm configurations, they
can overwhelm traditional naval defenses, saturate enemy sensors and weapon
systems, and complicate targeting decisions. Changing the calculus of engage-
ment, especially in coastal environments or during rapid-response operations.
2. Covert Operations and Sabotage
Maritime drones enable new forms of covert action. Their small size and stealth
features allow them to infiltrate harbors, lay mines, and damage critical infras-
tructure such as undersea cables or energy pipelines. These operations can be
carried out with minimal risk to personnel and may remain undetected until
the effects are noticed. Opening new possibilities for sabotage, subversion, and
strategic disruption, especially when attribution is uncertain or contested.
3. Logistics and Resupply
In addition to traditional combat applications, maritime drones are being
6.3 EMERGING TECHNOLOGIES AND IMPROVEMENTS 64
adapted for logistical tasks. Maritime drones can transport supplies, am-
munition, and equipment to deployed forces, reducing the risk to traditional
vessels and supporting sustained operations. Additionally, some drones are
designed for autonomous search-and-rescue missions or recovery operations in
hazardous conditions.
4. Counter-Drone Operations
Increased use of maritime drones creates a need for new defensive measures.
Various parties are currently developing counter-drone systems, including elec-
tronic warfare techniques, directed energy weapons, and drone-on-drone inter-
ception strategies. As these systems become increasingly interconnected and
reliant on software, cybersecurity rises as a critical vulnerability. A compro-
mised drone system could mislead, disable, or turn against its fleet.
6.3 Emerging Technologies and Improvements
Looking at the innovations from Chapter 4, we can observe that maritime drones
have undergone a lengthy process of improvement in terms of efficiency and modular
payload capabilities. This section will answer the second research question: 2. How
can maritime drones be optimized and equipped better?
The maritime robotics field is currently experiencing rapid growth, with innova-
tions emerging every month. These advancements have taken the world by surprise
in various ways. For instance, on the military side, traditional fighter jet carriers
are being replaced by fully autonomous drone carriers able to transport hundreds of
drones. Additionally, there are motherships designed to contain smaller drones that
can be deployed for their missions, as well as connected drone swarms.
A notable aspect of developing military applications is the modularity of these
systems, allowing them to be equipped with almost any payload. Military appli-
6.3 EMERGING TECHNOLOGIES AND IMPROVEMENTS 65
cations require a variety of payloads, ranging from traditional to electronic warfare
payloads. The diverse range of maritime drones presented provided us with a clear
view of these capabilities.
Maritime drones are increasingly being outfitted with advanced weapon systems
to enhance their capabilities to the fullest extent. An excellent example is one of the
newest maritime drones from Ukraine, the Katran Venom. This drone is equipped
with two torpedoes, two air defense missiles, and two machine guns: a 7.62-caliber
minigun and a 50-caliber Browning M2 machine gun. Additionally, it includes elec-
tronic warfare equipment and smoke and heat decoy systems. Furthermore, Katran
can carry a range of additional payloads, including aerial drones and various weapon
systems. Katran Venom is shown in Figure 6.1. [65]
Figure 6.1: Ukraine’s maritime drone, the Katran Venom. [65]
On the efficiency side, maritime drones are being optimized with innovative
designs that prioritize hydrodynamic performance, incorporate more efficient and
alternative battery technologies and advanced software, improve power management
6.4 ETHICAL AND ENVIRONMENTAL CONCERNS 66
systems, and integrate better sensors.
These enhancements will increase the capabilities of maritime drones, enabling
them to operate more effectively in swarms and to carry a greater arsenal of lethal
and non-lethal weapon systems. Enable longer operational time and range. To
predict the future, multi-domain robot attacks are likely the next wave of operations
to be seen, involving synchronized autonomous or unmanned attacks using ground,
air, and sea drones simultaneously.
6.4 Ethical and Environmental Concerns
This and the following sections will answer the third research question: 3. What
are the ethical and environmental considerations surrounding the deploy-
ment of maritime drones? And what kind of regulatory measures have
been set for maritime drones?
Maritime drones have already caused ethical and security concerns and will con-
tinue to do so in the future. In the article published in the Ukrainian open-access
law scientific journal Maritime Autonomous Weapon Systems from the Standpoint
of International Humanitarian Law, possible ethical impacts of maritime drones
in the view of International Humanitarian Law (IHL) are presented. Usmanov and
Chernychka explain that the main reason why this topic is debated is that the use of
lethal force is moving more toward autonomy. Therefore, meaning that autonomous
weapon systems can remove humans completely from the loop of decision. Usmanov
and Chernycka present legal issues that the use of autonomous weapon systems can
have even though we have seen deployment of such systems from various countries.
[38]
The most known and debated issue that is presented is the ethical concern of
delegating life-and-death decisions to autonomous machines. This issue is universal
for all autonomous machines that can be responsible for human life. [38] Who will
6.4 ETHICAL AND ENVIRONMENTAL CONCERNS 67
take the blame for killing the human or destroying the asset? Would it be the
operator who launched it, the company that built it, or even a single programmer
who made the algorithm to kill and destroy? Furthermore, to continue on this
topic, a critical point is that these systems need to distinguish between civilians
and combatants, and by doing so, they should try to do so in a way that produces
minimal harm since it is guaranteed that in a total war situation, there will be
civilian casualties. And what to do if a civilian is using some uniform that can
be mistaken for a military uniform? Additionally, there is a large issue with the
definitions and classification of these systems, and therefore it is hard to establish
new regulations and oversight on AWS. [38]
Regarding general security, these systems can both enhance and undermine it,
particularly due to privacy concerns [38]. Maritime drones are equipped with mul-
tiple sensors, such as cameras, and may record individuals without their consent.
This raises important legal and ethical questions: while it is generally legal to film
people in public spaces, recording private property or tracking someone’s movements
can be more problematic. So, where should we draw the line?
On the environmental side, maritime drones are pretty unnoticeable. However,
the greater issue is explosive-laden military maritime drones that scatter pieces
all around the area of explosion; this can be anything from oil, gasoline, plastics,
electronics, explosive materials, et cetera. Everyone knows this is an issue, but
nothing can really be done to minimize it other than forbidding all of them. On the
other hand, there are maritime drones that are designed to clean the oceans, so this
is an endless cat-and-mouse race. Furthermore, maritime drones can affect sea life
in a negative way, since the noise and debris can affect sensitive animals and other
sea life.
6.5 POLICY AND REGULATORY CONSIDERATIONS 68
6.5 Policy and Regulatory Considerations
The regulation of AWS remains an evolving field. The first step toward global regu-
lation of such technologies was taken in 2019 when a set of eleven guiding principles
was introduced by the Governmental Expert Group (GGE). These principles marked
the initial attempt to provide a normative framework for the development and de-
ployment of lethal AWS. However, many countries have not adopted it. Usmanov
and Chernychka explain that there is no need to ban autonomous weapons since
they do not cause unnecessary suffering. The stance of maritime drones within the
framework of international humanitarian law is that they can only be subject to it
if they operate under human control and decision-making. [38]
The Geneva Conventions have historically functioned as the foundation of inter-
national humanitarian law during armed conflict. Applying it to emerging maritime
drones presents significant challenges. For example, in order to fall under the exist-
ing wartime legal framework, they would first need to be classified as ships. Only
after this initial categorization could they potentially be recognized as warships, a
designation that comes with stringent legal and structural requirements that most
autonomous maritime systems do not meet. [66]
One illustrative example of this legal complexity is found in Article 18 of the
Second Geneva Convention, which requires parties of a conflict to actively search
for and assist the wounded, shipwrecked, and sick after naval engagements:
1. "Article 18 – Search for casualties after an engagement: After each
engagement, Parties to the conflict shall, without delay, take all possible mea-
sures to search for and collect the shipwrecked, wounded and sick, to protect
them against pillage and ill-treatment, to ensure their adequate care, and to
search for the dead and prevent their being despoiled." [67]
This becomes operationally impractical in the context of small, autonomous mar-
6.5 POLICY AND REGULATORY CONSIDERATIONS 69
itime drones, particularly those designed for kamikaze-style attacks, which lack any
capability for rescue or post-conflict engagement. While there is ongoing research
and development into specialized unmanned systems intended for maritime search
and rescue, these technologies have not yet matured to a stage where they can be
reliably deployed in real-world operations.
Among the existing legal instruments, Article 36 of the First Additional Protocol
to the Geneva Conventions may be the most applicable to autonomous maritime
systems. This provision mandates that each state must conduct legal reviews of
new weapons to ensure compliance with international law:
1. "Article 36 – New weapons: In the study, development, acquisition or
adoption of a new weapon, means or method of warfare, a High Contracting
Party is under an obligation to determine whether its employment would, in
some or all circumstances, be prohibited by this Protocol or by any other rule
of international law applicable to the High Contracting Party." [67]
This article creates a legal obligation for states to assess whether new tech-
nologies, including maritime drones, conform to existing laws and norms before
deployment. However, the application of Article 36 is still ambiguous in the context
of autonomous systems, as it leaves significant room for interpretation regarding
what constitutes a new method of warfare and whether current legal standards are
sufficient.
6.5.1 European Union
While international frameworks such as the Geneva Conventions and the previously
mentioned guiding principles offer general reference points, they are insufficient for
comprehensively regulating autonomous maritime systems. Consequently, the Euro-
pean Union and individual states have begun to explore region-specific and national
6.5 POLICY AND REGULATORY CONSIDERATIONS 70
regulatory approaches to address the legal and ethical challenges posed by these
technologies.
In 2021, the European Commission introduced the Artificial Intelligence Act
(AI Act). It is a proposed legal framework aimed at ensuring trustworthy and
human-centric AI development across member states. While primarily focused on
civilian applications, the regulation identifies high-risk AI systems and encourages
transparency, oversight, and accountability mechanisms. [68] It does not specifically
take into account military applications. However, it can have a negative effect during
the development of dual-use applications.
Furthermore, the EU Code of Conduct for Artificial Intelligence in Defense,
published by the European Defence Agency (EDA) in May 2025, outlines voluntary
commitments for the ethical use of AI in military contexts. This code emphasizes
aspects such as human responsibility, accountability, and compliance with interna-
tional humanitarian law. Although non-binding, it shows the EU’s recognition of
the risks associated with integrating AI into autonomous systems. Additionally, it is
quite broad since it takes different levels of warfare into account and how AI should
be used at given levels. [69]
6.5.2 National-Level Regulations
On the national level, regulatory responses vary widely depending on each country’s
technological capabilities, defense policies, and legal traditions. For instance, looking
at a few of the larger military powers [70]:
Germany has taken a cautious approach, consistently advocating for interna-
tional bans or strict regulation of lethal autonomous weapon systems. The German
government supports the principle that human decision-making must remain central
in the use of force, aligning closely with ethical constraints grounded in international
humanitarian law.
6.5 POLICY AND REGULATORY CONSIDERATIONS 71
France has supported further development of military AI but also stresses the
importance of human oversight. French defense doctrine emphasizes meaningful
human control over weapon systems, and the country has actively participated in
international forums discussing the regulation of LAWS. Additionally, China and
Russia highlight the principle of meaningful human control in AWS but are quite
permissive with the development of these systems.
The United Kingdom and the United States, on the other hand, have shown a
more permissive stance, emphasizing innovation and operational effectiveness. While
the UK upholds international legal obligations, it has not endorsed proposals for
a binding international ban on autonomous weapons. British policy documents
often refer to autonomous systems within a framework that prioritizes their military
utility. In the U.S., the Department of Defense Directive 3000.09 is a regulation for
AWS. The original was established in 2012 and revised in 2023. The U.S. has
quite a different stance compared to others since they state that AWS should have
appropriate levels of human judgment, which can be interpreted very differently from
meaningful human control or human control. The 2012 version of this directive is
one of the first-ever national-level regulations for AWS. [71] This directive allows
the U.S. to develop AWS that may require no more than human supervision.
These national-level efforts reflect a fragmented regulatory environment, where
states adopt different interpretations of their legal obligations and ethical respon-
sibilities. This inconsistency complicates efforts to develop a harmonized global or
even regional framework. The regulation report [70] states that 195 countries have
issued statements on AWS. Of these, 129 support regulating AWS, while 12 oppose
such regulation. The remaining 54 have not taken a clear stance. This means that
approximately 66% of the countries favor regulating AWS.
In conclusion, there is currently no unified or comprehensive legal framework that
adequately addresses the classification and regulation of maritime drones. Their
6.5 POLICY AND REGULATORY CONSIDERATIONS 72
varied sizes, functionalities, and intended uses ranging from reconnaissance to di-
rect engagement or search and rescue complicate attempts at universal categoriza-
tion. However, some similarities can be seen, especially in the aspect of keeping the
decision-making in the hands of humans.
7 Conclusion
This chapter concludes the thesis by summarizing the work completed and outlining
potential directions for future research.
7.1 Summary
This thesis explored methods to enhance the efficiency of maritime drones and ex-
amined emerging trends and innovations in the future of naval drones. It examined
the evolution of maritime drones and their development from the early days of World
War I to the modern 21st century, with warfare as the key focus point.
The thesis introduced popular robotics tools for developing maritime drones,
such as Robot Operating System and Mission Oriented Operating Suite Interval
Programming. This thesis also explored various simulators and core technical foun-
dations for developing maritime drones, along with current challenges, the global
market, and future forecasts and estimations.
This thesis addresses the defined research questions, which include examining
the tactical and strategic impacts of maritime drones in modern warfare, exploring
how they can be better equipped and optimized, identifying the environmental and
ethical challenges they pose, and reviewing the current regulatory frameworks.
The alarming reality of these military systems showcases the need for improved
countermeasures, more advanced drones, and further innovations. The importance
of unmanned naval systems is now evident. They serve as cost-effective weapons
7.2 FUTURE WORK 74
on multiple levels of warfare. Nations with even no naval force can rely on a mar-
itime drone fleet to attack enemy vessels. And we are just at the beginning of
these systems. Initially, we saw only kamikaze drones. Currently, maritime drones
can be equipped with air support, additional drones, torpedoes, electronic warfare
equipment, and small arms. Maritime drones represent a transformative shift in
naval warfare, with implications that extend from the tactical battlefield to the
global strategic environment. Their affordability, flexibility, and autonomy make
them a disruptive force that empowers new actors, enables covert operations, and
challenges established doctrines of naval power. As different parties adapt to these
technologies, the maritime domain is assured to become more contested, automated,
and complex. The implications for deterrence, escalation, and global security are
profound and will only grow as maritime drones continue to evolve.
7.2 Future Work
As mentioned earlier, marine robotics is not as advanced as other fields of robotics.
Therefore, governments and private companies should invest more in these systems,
as we have seen them successfully applied in real-life scenarios. Further research
and innovation in maritime drone technology are essential on a global scale. This is
especially true for Finland, which remains significantly behind other Nordic nations
in advancing this sector. It is my perspective that the Finnish Defense Forces
could contribute to this development far more without interfering with their regular
activities. Collaborating with Ukraine has already led to some advancements in the
field. The use of this technology in Finland is not new, as the Finnish Navy has
employed it, but on a limited scale. Additionally, maritime drones developed in
Western nations should adhere to NATO and EU standards. This standardization
is crucial for guaranteed interoperability among maritime drone systems, which is
vital for the collective defense of Western countries.
7.2 FUTURE WORK 75
Furthermore, all developed maritime drones should have dual-use capabilities,
allowing them to be modular and accommodate both military and civilian payloads.
Currently, the largest joint military exercise on maritime drones is Robotic Ex-
perimentation and Prototyping Maritime Unmanned Systems Exercise (REPMUS),
which takes place annually. In the REPMUS 2024 exercise, 2000 participants from
30 NATO and partner countries participated [72]. The next REPMUS will take
place on the 25th of September, 2025, and participants are expected to demonstrate
an even wider range of new maritime capabilities [73]. This joint exercise is a clear
sign of the increased development and interest in these drones. The first fully au-
tonomous war has yet to occur, but each technological advancement brings us closer
to that reality.
References
[1] D. Axe, “Ukraine’s All-Robot Assault Force Just Won Its First Battle”, en,
Forbes, 2024. [Online]. Available: https://www.forbes.com/sites/davidax
e/2024/12/21/ukraines-first-all-robot-assault-force-just-won-it
s-first-battle/ (visited on 05/30/2025).
[2] A. Fratsyvir, “Ukraine’s AI-powered ‘mother drone’ sees first combat use, min-
ister says”, en, The Kyiv Independent, 2025. [Online]. Available: https://kyi
vindependent.com/ukraines-ai-powered-mother-drone-sees-first-co
mbat-use-minister-says/ (visited on 05/30/2025).
[3] Unmanned surface vehicle - USV, en. [Online]. Available: https://en.wiki
pedia.org/w/index.php?title=Unmanned_surface_vehicle (visited on
05/25/2025).
[4] Unmanned underwater vehicle - UUV, en. [Online]. Available: https://en.wi
kipedia.org/w/index.php?title=Unmanned_underwater_vehicle (visited
on 05/25/2025).
[5] O. Danylov, “Naval strike drones: From ancient brigands to modern times”,
en, Mezha.Media, 2023. [Online]. Available: https://mezha.media/en/arti
cles/naval-strike-drones-from-ancient-brigands-to-modern-times/
(visited on 03/17/2025).
[6] J. Saul and R. Maltezou, “Houthi explosive drone boat attacks escalate Red
Sea danger”, en, Reuters, 2024. [Online]. Available: https://www.reuters.c
REFERENCES 77
om/world/middle-east/houthi-explosive-drone-boat-attacks-escala
te-red-sea-danger-2024-07-03/ (visited on 05/21/2025).
[7] J. Leman, “Tekoälyavusteiset asejärjestelmät nykyaikaisessa sodankäynnissä”,
fi, 2023. [Online]. Available: https://www.utupub.fi/handle/10024/179850
(visited on 05/25/2025).
[8] Remotely operated underwater vehicle, en. [Online]. Available: https://en.w
ikipedia.org/w/index.php?title=Remotely_operated_underwater_vehi
cle (visited on 05/25/2025).
[9] Nato marine unmanned systems, en, 2020. [Online]. Available: https://www
.nato.int/nato_static_fl2014/assets/pdf/2020/11/pdf/2011-facthee
t-mus.pdf (visited on 05/25/2025).
[10] European defense agency - UMS, en. [Online]. Available: https://eda.europ
a.eu/what-we-do/all-activities/activities-search/unmanned-marit
ime-systems-(ums)-research (visited on 05/16/2025).
[11] P. Small, Unmanned maritime systems, en, 2019. [Online]. Available: https:
//www.navsea.navy.mil/Portals/103/Documents/Exhibits/SNA2019/Un
mannedMaritimeSys-Small.pdf?ver=2019-01-15-165105-297 (visited on
05/26/2025).
[12] T. Nikola, “Method of and apparatus for controlling mechanism of moving
vessels or vehicles”, en, US613809A, 1898. [Online]. Available: https://pate
nts.google.com/patent/US613809/en (visited on 05/25/2025).
[13] FL-boat, en. [Online]. Available: https://en.wikipedia.org/w/index.php
?title=FL-boat (visited on 05/25/2025).
[14] Protector USV, en. [Online]. Available: https://en.wikipedia.org/w/inde
x.php?title=Protector_USV (visited on 05/25/2025).
REFERENCES 78
[15] “ACTUV “Sea Hunter” Prototype Transitions to Office of Naval Research for
Further Development”, en, 2018. [Online]. Available: https://www.darpa.mi
l/news/2018/sea-hunter-prototype (visited on 05/27/2025).
[16] T. Ozberk, “MARLIN USV: Meet Turkiye’s Latest Drone Ship”, en, Naval
News, 2022. [Online]. Available: https://www.navalnews.com/naval-news
/2022/09/marlin-usv-meet-turkiyes-latest-drone-ship/ (visited on
05/21/2025).
[17] A. Mazurenko, “Russia creates marine drone similar to Ukrainian SeaBaby”,
en, Ukrainska Pravda, 2023. [Online]. Available: https://www.pravda.com.u
a/eng/news/2023/12/15/7433230/ (visited on 05/25/2025).
[18] H. Sutton, Russia faces new threat: Ukraine’s ‘Toloka’ underwater maritime
drone, en, 2023. [Online]. Available: http://www.hisutton.com/New-Ukrain
e-Underwater-Maritime-Drone.html (visited on 05/22/2025).
[19] H. R. Widditsch, “SPURV - The First Decade:” en, Defense Technical Informa-
tion Center, Fort Belvoir, VA, Tech. Rep., 1973. doi: 10.21236/ADA050816.
(visited on 05/26/2025).
[20] A. Bursuc et al., “Overview on Sea Drones Evolution and their Use in Modern
Warfare”, en, Land Forces Academy Review, vol. 29, no. 2, pp. 195–209, 2024.
doi: 10.2478/raft-2024-0021. (visited on 03/27/2025).
[21] History of UUV | Exploring NUSTEM, en. [Online]. Available: https://nus
tem.bridgeport.edu/history-of-uuv/ (visited on 05/26/2025).
[22] H. Sutton, Captured UUV in Pyongyang North Korea, en, 2020. [Online]. Avail-
able: http://www.hisutton.com/Captured-UUV-in-Pyongyang-North-Kor
ea.html (visited on 06/02/2025).
REFERENCES 79
[23] E. Ciobanu, The History of Underwater Drones - Droneblog, en, 2018. [Online].
Available: https://www.droneblog.com/the-history-of-underwater-dro
nes/ (visited on 06/02/2025).
[24] REMUS (vehicle), en. [Online]. Available: https://en.wikipedia.org/w/in
dex.php?title=REMUS_(vehicle) (visited on 06/02/2025).
[25] Manta Ray, en. [Online]. Available: https:www.northropgrumman.com/what
-we-do/sea/manta-ray (visited on 04/14/2025).
[26] Anduril industries, inc. en. [Online]. Available: https://www.anduril.com
(visited on 05/25/2025).
[27] S. Gujar and P. Paul, “Unmanned marine vehicles market size, share & analysis
– 2034”, en, Global Market Insights Inc., 2024. [Online]. Available: https://w
ww.gminsights.com/industry-analysis/unmanned-marine-vehicles-ma
rket (visited on 04/11/2025).
[28] A. Miller, B. Miller, and G. Miller, “Navigation of Underwater Drones and
Integration of Acoustic Sensing with Onboard Inertial Navigation System”,
en, Drones, vol. 5, no. 3, p. 83, 2021. doi: 10.3390/drones5030083. (visited
on 03/10/2025).
[29] E. Zereik et al., “Challenges and future trends in marine robotics”, en, Annual
Reviews in Control, vol. 46, pp. 350–368, 2018. doi: 10.1016/j.arcontrol.2
018.10.002. (visited on 05/30/2025).
[30] Y. Wu et al., “A Review of Path Planning Methods for Marine Autonomous
Surface Vehicles”, en, Journal of Marine Science and Engineering, vol. 12,
no. 5, p. 833, 2024. doi: 10.3390/jmse12050833. (visited on 03/09/2025).
[31] F. Rubino et al., “Marine Application of Fiber Reinforced Composites: A Re-
view”, en, Journal of Marine Science and Engineering, vol. 8, no. 1, p. 26,
2020. doi: 10.3390/jmse8010026. (visited on 06/01/2025).
REFERENCES 80
[32] ROS 2 Documentation: Kilted documentation, en. [Online]. Available: https:
//docs.ros.org/en/kilted/index.html (visited on 05/27/2025).
[33] MOOS-IvP Documentation, en. [Online]. Available: https://oceanai.mit.e
du/moos-ivp (visited on 06/01/2025).
[34] M. E. Orye et al., “Enhancing the Cyber Resilience of Sea Drones”, en, in 2024
16th International Conference on Cyber Conflict: Over the Horizon (CyCon),
Tallinn, Estonia: IEEE, 2024, pp. 83–102. doi: 10.23919/CyCon62501.2024
.10685581. (visited on 06/01/2025).
[35] E. Jokioinen et al., “Rolls Royce AAWA Report”, en, Tech. Rep., 2016. [On-
line]. Available: https://www.utupub.fi/handle/10024/157117 (visited on
05/21/2025).
[36] T. McMillan, Stealth Technology Breakthrough By Navy Researchers Could
Revolutionize Underwater Drone Capabilities, en, 2025. [Online]. Available:
https://thedebrief.org/stealth-technology-breakthrough-by-navy
-researchers-could-revolutionize-underwater-drone-capabilities/
(visited on 06/01/2025).
[37] P. Giangiuseppe, Sea drones at war: Tactical, operational and strategic analysis
of maritime uncrewed systems, en, 2024. [Online]. Available: https://euro-
sd.com/2024/09/articles/40191 (visited on 03/18/2025).
[38] I. Usmanov and M. Chernychka, “Maritime autonomous weapon systems from
the standpoint of international humanitarian law”, en, Lex Portus, 2022. doi:
10.26886/2524-101X.8.2.2022.2. (visited on 03/20/2025).
[39] Fluid dynamics, en. [Online]. Available: https://en.wikipedia.org/w/inde
x.php?title=Fluid_dynamics (visited on 05/27/2025).
REFERENCES 81
[40] The Wave Glider | Unmanned Surface Vehicle by Liquid Robotics, en. [Online].
Available: https://www.liquid-robotics.com/wave-glider/overview/
(visited on 04/14/2025).
[41] N. Rennolds, “Huge US military ’Manta Ray’ sea drone spotted on Google
Earth at California naval base”, en, Business Insider, 2024. [Online]. Available:
https://www.businessinsider.com/us-navy-manta-ray-sea-drone-cal
ifornia-coast-google-earth-2024-6 (visited on 04/14/2025).
[42] The TRITON, en. [Online]. Available: https://www.oceanaero.com/the-tr
iton (visited on 04/14/2025).
[43] M. Tanner, “Tällainen on Suomessa kehitetty moottoroitu ja aseistettava tais-
telusurffilauta "Meri-Turso"”, fi, Tekniikka&Talous, 2024. [Online]. Available:
https://www.tekniikkatalous.fi/uutiset/tallainen-on-suomessa-keh
itetty-moottoroitu-ja-aseistettava-taistelusurffilauta-meri-tur
so-jolla-voisi-valvoa-venalaisten-sukellusveneita/6821d8ae-4ed4
-4cf8-99a0-9ae7600d374a (visited on 03/31/2025).
[44] V. Kykkänen, “Näin Suomessa kehitetään nyt drooniteknologiaa sotaan: "siir-
rytään maratonista pikamatkoille"”, fi, mtvuutiset.fi, 2025. [Online]. Available:
https://www.mtvuutiset.fi/artikkeli/nain-suomessa-kehitetaan-ny
t-drooniteknologiaa-sotaan-siirrytaan-maratonista-pikamatkoille
/9091648 (visited on 03/31/2025).
[45] J. Rice-Jones and A. Pande, “Best single board computers in 2025”, en, XDA,
2023. [Online]. Available: https://www.xda-developers.com/best-single
-board-computer/ (visited on 06/02/2025).
[46] ADAPTIV - Cloak of Invisibility, en. [Online]. Available: https://www.baes
ystems.com/en-us/product/adaptiv (visited on 06/02/2025).
REFERENCES 82
[47] C. H. Popa et al., “Distributed Maritime Operations and Unmanned Systems
Tactical Employment”, en, Tech. Rep., 2018. [Online]. Available: https://ca
lhoun.nps.edu/bitstream/handle/10945/59587/18Jun_SE_Capstone_Pop
a_et_al.pdf (visited on 06/02/2025).
[48] H. Uçar, “RADAR CROSS SECTION REDUCTION”, en, Journal of Naval
Science and Engineering, vol. 9, no. 2, pp. 72–87, 2013. [Online]. Available:
https://www.researchgate.net/publication/305721036_Radar_Cross
_Section_Reduction (visited on 06/02/2025).
[49] C.-h. Hung and J. Chung, “Uncrewed surface vehicle unveiled - Taipei Times”,
en, 2025. [Online]. Available: https://www.taipeitimes.com/News/front/a
rchives/2025/03/26/2003834074 (visited on 06/02/2025).
[50] M. El-Hajj, “Enhancing Communication Networks in the New Era with Arti-
ficial Intelligence: Techniques, Applications, and Future Directions”, en, Net-
work, vol. 5, no. 1, p. 1, 2025. doi: 10.3390/network5010001. (visited on
06/02/2025).
[51] P. Giordano, “Dynamic Messenger 23 Provides Lessons For Autonomous Ve-
hicles in NATO Multi Domain Operations”, en, NATO’s ACT, 2023. [Online].
Available: https://www.act.nato.int/article/dynmsg23-furthers-nato
-mdo/ (visited on 06/02/2025).
[52] M. Zetas et al., “Empowering 6G maritime communications with distributed
intelligence and over-the-air model sharing”, en, Frontiers in Communications
and Networks, vol. 4, 2024. doi: 10.3389/frcmn.2023.1280602. (visited on
06/06/2025).
[53] 385th Unmanned Surface Vehicles Brigade (Ukraine), en. [Online]. Available:
https://en.wikipedia.org/w/index.php?title=385th_Unmanned_Surfac
e_Vehicles_Brigade_(Ukraine) (visited on 06/06/2025).
REFERENCES 83
[54] A. Fratsyvir, “Ukrainian drones destroy Russian radar, supply depots on Black
Sea gas platforms, SBU says”, en, The Kyiv Independent, 2025. [Online]. Avail-
able: https://kyivindependent.com/ukrainian-drones-destroy-russi
an- radar- and- supply- depots- on- black- sea- drilling- platforms/
(visited on 05/27/2025).
[55] J. Waterhouse and K. Armstrong, “Russian ship hit in Novorossiysk, Black
Sea drone attack, Ukraine sources say”, en, 2023. [Online]. Available: https:
//www.bbc.com/news/world-europe-66402046 (visited on 04/17/2025).
[56] H. Sutton, Overview of maritime drones (usvs) of the russo-ukrainian war,
2022-24, en, 2024. [Online]. Available: http://www.hisutton.com/Russia-
Ukraine-USVs-2024.html (visited on 05/21/2025).
[57] M. Bubalo and E. Goksedef, “Ukraine says it launched July attack on bridge
to Crimea”, en, 2023. [Online]. Available: https://www.bbc.com/news/worl
d-europe-66397227 (visited on 04/15/2025).
[58] S. Scarr et al., “How Yemen’s Houthi rebels are carrying out attacks on Red
Sea ships”, en, Reuters, 2024. [Online]. Available: https://www.reuters.co
m/graphics/ISRAEL-PALESTINIANS/SHIPPING-ARMS/lgvdnngeyvo/ (visited
on 04/20/2025).
[59] E. Klutstein, The new front against iran and its proxies: Underwater, en, 2024.
[Online]. Available: https://www.misgavins.org/en/klutstein- the-
new- front- against- iran- and- its- proxies- underwater/ (visited on
04/20/2025).
[60] H. Sutton, Houthi’s blowfish: Guide to explosive USV threat in Red Sea, en,
2024. [Online]. Available: http://www.hisutton.com/Yemen-Houthi-USV-
Guide.html (visited on 05/21/2025).
REFERENCES 84
[61] BlueWhale (UUV), en. [Online]. Available: https://en.wikipedia.org/w/i
ndex.php?title=BlueWhale_(UUV) (visited on 05/27/2025).
[62] I. Hilton, “Taiwan Makes the Majority of the World’s Computer Chips. Now
It’s Running Out of Electricity”, en, Wired, [Online]. Available: https://w
ww.wired.com/story/taiwan- makes- the- majority- of- the- worlds-
computer- chips- now- its- running- out- of- electricity/ (visited on
06/03/2025).
[63] C. T. Moon, Chinese Seize U.S. Navy Underwater Drone in South China Sea,
en. [Online]. Available: https://www.defense.gov/News/News-Stories/Ar
ticle/Article/1032823/ (visited on 06/03/2025).
[64] M. Hernandez, “Maps: Tracking Tensions Between China and Taiwan”, en,
The New York Times, 2022. [Online]. Available: https://www.nytimes.co
m/interactive/2022/world/asia/taiwan-china-maps.html (visited on
05/21/2025).
[65] KATRAN "VENOM" - MAC - Military Armored Company HUB, uk, 2024.
[Online]. Available: https : / / vpk - mac . com / katran - venom/ (visited on
04/14/2025).
[66] C. M. Liu, The Unresolved Legal Status of Maritime Drones: (War)ships or
Weapons? - RSIS, en, 2024. [Online]. Available: https://rsis.edu.sg/rsis
-publication/idss/ip24097 (visited on 06/09/2025).
[67] The Geneva Conventions and their Commentaries | ICRC, en, 2024. [Online].
Available: https://www.icrc.org/en/law-and-policy/geneva-conventio
ns-and-their-commentaries (visited on 06/11/2025).
[68] European Union AI Act: First regulation on artificial intelligence, en, 2023.
[Online]. Available: https://www.europarl.europa.eu/topics/en/articl
e/20230601STO93804 (visited on 06/09/2025).
REFERENCES 85
[69] Trustworthiness for ai in defence, en, 2025. [Online]. Available: https://eda
.europa.eu/docs/default-source/brochures/taid-white-paper-final
-09052025.pdf (visited on 06/09/2025).
[70] T. Jones, How Nations are Responding to Autonomous Weapons in War, en,
2025. [Online]. Available: https://autonomousweapons.org/global-persp
ectives-on-regulation/ (visited on 06/10/2025).
[71] K. M. Sayler, Defense Primer: U.S. Policy on Lethal Autonomous Weapon
Systems, en, legislation, 2025. [Online]. Available: https://www.congress.g
ov/crs-product/IF11150 (visited on 06/10/2025).
[72] NATO, NATO’s Digital Ocean Initiative gets a boost in Portugal, en, 2024.
[Online]. Available: https://www.nato.int/cps/en/natohq/news_228959.h
tm (visited on 05/21/2025).
[73] REPMUS 2025 CALLING NOTICE - Find a Tender, en, 2025. [Online]. Avail-
able: https://www.find-tender.service.gov.uk/Notice/007342-2025
(visited on 05/21/2025).