Environment. Technology. Resources.

Rezekne, Latvia
Proceedings of the 15th International Scientific and Practical Conference. Volume IV, 138-143

AI-Enabled Drone Autonomous Navigation and

Decision Making For Defence Security
Amit Joshi
Doctor of Business Management
BA School of Business and Finance
Riga-Latvia
amit.joshi00008@gmail.com
Aivars Spilbergs Elīna Miķelsone
Doctor of Business Management Doctor of Business Management
BA School of Business and BA School of Business and Finance
Finance Riga-Latvia
Riga-Latvia mikelsone.elina@gmail.com
aivars.spilbergs@ba.lv

Abstract. The combination of Artificial Intelligence (AI) and
unmanned aerial vehicles (UAVs), sometimes known as
drones, has become a revolutionary approach in modern
military and security operations. The purpose of this study
is to explore and assess the efficiency of AI-enabled
autonomous navigation and decision-making systems for
drones in defense security applications. Through a
comprehensive literature review, researchers analyze the
various AI techniques and algorithms used in these systems,
including machine learning, deep learning, and
reinforcement learning. The study examines different
aspects of autonomous drone navigation, such as sensors,
decision-making modules, communication systems, and
countermeasure systems. By reviewing scholarly articles
and existing studies, researchers gain insights into the
hardware and software components, including GPS
modules, IMUs, cameras, and other sensors. This analysis
provides a clear understanding of the current state of AI-
enabled drone technology for defense security and identifies
potential areas for future research and improvement. This
research study discusses the working of AI-enabled drone
autonomous navigation and decision-making systems
designed primarily for defense security applications. The
study starts by explaining the structure of drone navigation
systems, which includes a wide range of hardware and
software components. These comprise GPS modules for
tracking location, inertial measurement units (IMUs) for
estimating attitude, and cameras for seeing the
environment. By incorporating these sensors into a sturdy
structure, drones are able to detect their surroundings and
man oeuvre independently in intricate situations. The
effectiveness of AI-enabled drone navigation relies heavily
on the application of sophisticated artificial intelligence
techniques and algorithms. Machine learning algorithms,
such as deep neural networks and reinforcement learning,
are crucial in improving the decision-making abilities of
drones. AI algorithms allow drones to dynamically adjust
their navigation tactics, optimize flight trajectories, and
intelligently respond to unforeseen obstacles or hazards by
analyzing large volumes of sensor data in real-time.
Furthermore, this research explores the datasets being
employed in the training and evaluation of AI models for
the purpose of drone navigation and decision-making. These
datasets contain varied environmental conditions,
topographical features, and security scenarios experienced
in defensive operations.
Keywords: Artificial Intelligence, Aerial Security, Defense
Technology and Drone Surveillance.
I. INTRODUCTION
The development of autonomous drone detection and
navigation systems represents notable progress in the field
of defence technology. The increasing prevalence of
unmanned aerial vehicles (UAVs) in areas such as
defence, security, and commercial applications has
created a growing demand for reliable systems that can
autonomously detect, track, and navigate drones.
Researchers and engineers have utilized artificial
intelligence (AI) technology to create advanced systems
that can independently identify and manoeuvre drones in
intricate surroundings, in response to this requirement.
Artificial intelligence is crucial in the development of
autonomous drone detection and navigation systems,
allowing unmanned aerial vehicles (UAVs) to function
autonomously and make intelligent choices in real-time.
Artificial intelligence algorithms, namely those utilizing
machine learning and deep learning methods, enable
drones to accurately assess their surroundings, detect
possible dangers, and manoeuvre through changing
situations with accuracy and effectiveness. Autonomous
drone detection and navigation systems can enhance their
effectiveness over time by utilizing AI technology. They
are capable of adjusting to varying environments,
acquiring knowledge from previous encounters, and
consistently refining their abilities. Autonomous drone
detection and navigation systems are used in defence
applications to defend important infrastructure, monitor
borders, and safeguard military facilities against aerial

Print ISSN 1691-5402

Online ISSN 2256-070X
https://doi.org/10.17770/etr2024vol4.8237
© 2024 Amit Joshi, Aivars Spilbergs, Elīna Miķelsone.
Published by Rezekne Academy of Technologies.
This is an open access article under the Creative Commons Attribution 4.0 International License

138
Amit Joshi. et al. AI-Enabled Drone Autonomous Navigation and Decision Making For Defence Security
attacks. These technologies are crucial for improving
situational awareness, allowing defence personnel to
identify unauthorized drones and promptly take
appropriate action to eliminate any risks. The purpose of
this study is to develop and understand the efficiency of
artificial intelligence (AI)-enabled autonomous navigation
and decision-making systems for drones, particularly for
defence security applications. The study analyzes different
aspects and functioning of autonomous navigation and
decision-making systems for drones that are primarily
designed for defence security applications. The research
task involves studying the various AI techniques and
algorithms used in autonomous navigation and decision-
making systems. Researchers will also analyze the
hardware and software components used in these systems,
including GPS modules, inertial measurement units
(IMUs), cameras, and other sensor. The study utilizes a
comprehensive literature review to examine various AI
techniques and methodologies employed in drone
autonomous navigation and decision-making systems,
particularly in defence security applications. This
approach involves analyzing scholarly articles, research
papers, and existing studies that discuss and evaluate
different AI algorithms, including machine learning, deep
learning, and reinforcement learning, as well as their
applications in drone navigation. The research focuses on
the main components and functioning of autonomous
drone navigation systems, such as sensors, AI algorithms,
decision-making modules, communication systems, and
countermeasure systems. The review covers a range of
topics, including the use of computer vision algorithms for
object detection and recognition, machine learning
algorithms for pattern recognition and anomaly detection,
and deep learning algorithms for complex data analysis
and decision-making. By analyzing these studies,
researchers gain insights into the strengths, limitations,
and potential areas for improvement in AI-enabled drone
navigation and decision-making systems. This thorough
review helps provide a comprehensive understanding of
the current state of AI-enabled drone technology for
defense security and potential future research directions.
A. Main Components and Functioning
● Sensors: Sensors serve as the primary input
source for autonomous drone detection and
navigation systems, capturing data from the
drone's surroundings and providing vital
information for decision-making. Common
types of sensors used in these systems
include cameras, radar, lidar, and acoustic
sensors, each offering unique capabilities for
detecting and tracking drones in different
environments and conditions.
● AI Algorithms: AI algorithms form the
intelligence core of autonomous drone
detection and navigation systems, enabling
drones to perceive their environment,
interpret sensor data, and make informed
decisions autonomously. These algorithms
include computer vision algorithms for object
detection and recognition, machine learning
algorithms for pattern recognition and
anomaly detection, and deep learning
139
algorithms for complex data analysis and
decision-making.
● Decision-Making Modules: Decision-making
modules process sensor data and AI-
generated insights to make real-time
decisions regarding drone detection, tracking,
and response actions. These modules
incorporate rule-based logic, probabilistic
reasoning, and optimization techniques to
evaluate threats, assess risks, and determine
the appropriate course of action, such as
alerting operators, deploying
countermeasures, or initiating evasive
manoeuvres.
● Communication Systems: Communication
systems enable autonomous drone detection
and navigation systems to exchange data with
command and control centres, other drones,
and external sensors and platforms. These
systems utilize wireless communication
protocols, such as Wi-Fi, Bluetooth, and
cellular networks, to transmit sensor data,
status updates, and command instructions,
facilitating seamless coordination and
collaboration between multiple system
components and stakeholders.
● Countermeasure Systems: Countermeasure
systems provide autonomous drone detection
and navigation systems with the capability to
neutralize hostile drones and mitigate
potential threats. These systems include
electronic warfare techniques, such as
jamming and spoofing, physical interception
methods, such as net guns and drone-
capturing drones, and kinetic weapons, such
as lasers and missiles, each offering different
levels of effectiveness and precision in
countering aerial threats.
B. Functioning of Autonomous Drone Detection and
Navigation Systems
Autonomous drone detection and navigation systems
operate through a series of interconnected processes,
encompassing sensor data acquisition, AI-driven analysis,
decision-making, and response execution. The functioning
of these systems can be broadly categorized into the
following stages:
● Sensor Data Acquisition: The system's
sensors, including cameras, radar, lidar, and
acoustic sensors, continuously monitor the
drone's surroundings, capturing data related
to its position, velocity, trajectory, and
physical characteristics. These sensors detect
and track drones within the system's
operational range, providing real-time
situational awareness to the AI algorithms
● AI-Driven Analysis: AI algorithms analyze
sensor data to identify and classify potential
threats posed by drones, distinguishing
between authorized and unauthorized aerial
vehicles based on predefined criteria and
threat indicators. Computer vision algorithms
 Environment. Technology. Resources. Rezekne, Latvia
Proceedings of the 15th International Scientific and Practical Conference. Volume IV, 138-143
process visual data to detect drones in the
system's field of view, while machine
learning algorithms analyze patterns and
anomalies in sensor data to identify
suspicious behaviour.
● Decision-Making: Decision-making
modules evaluate AI-generated insights and
sensor data to make informed decisions
regarding threat assessment, response
prioritization, and action planning. These
modules incorporate rule-based logic,
probabilistic reasoning, and optimization
techniques to assess the severity of threats,
calculate response probabilities, and
determine the most effective course of action
based on predefined rules and objectives.
● Response Execution: Once a threat is
detected and assessed, the system initiates the
appropriate response actions to neutralize the
threat.
C. Artificial Intelligence Algorithms
Deep learning and machine learning algorithms are
crucial in the detection and decision-making processes of
autonomous drones. These algorithms include several
techniques such as convolutional neural networks
(CNNs), recurrent neural networks (RNNs), deep
reinforcement learning (DRL), generative adversarial
networks (GANs), and ensemble methods. Convolutional
Neural Networks (CNNs) are highly efficient in
performing image recognition tasks, allowing drones to
accurately detect and identify objects and obstacles in
their immediate environment. Recurrent Neural Networks
(RNNs), in contrast, are particularly suitable for analyzing
sequential data. Deep reinforcement learning algorithms
allow drones to acquire optimal strategies by engaging in
trial-and-error interactions with the environment, enabling
them to independently navigate intricate terrain and
prevent collisions. GANs enable the creation of lifelike
artificial data to train drone detection models, while
ensemble approaches merge different learning algorithms
to enhance overall performance and resilience. Through
the utilization of these sophisticated algorithms, self-
governing drones can attain improved detecting skills,
render well-informed decisions instantaneously, and
function efficiently in ever-changing and uncertain
surroundings.
In military applications, autonomous navigation and
decision-making for drones rely on a variety of AI
algorithms to ensure efficient and effective operation.
These algorithms, including Deep Reinforcement
Learning (DRL), Convolutional Neural Networks
(CNNs), Recurrent Neural Networks (RNNs), Fuzzy
Logic Systems, Bayesian Networks, Evolutionary
Algorithms, and Model Predictive Control (MPC), play
critical roles in enhancing drone capabilities.
D. Review of Literature
The literature on autonomous navigation and decision-
making for unmanned aerial vehicles (UAVs) covers a
wide range of research that investigate different elements
of UAV technology and its uses. Smith and Jones (2018)
present an extensive analysis of recent progress in
140
autonomous navigation, emphasizing the development of
navigation methods and the difficulties in guaranteeing
secure and efficient UAV operations. The conversation
revolves around the development of sophisticated
algorithms and sensor technologies that allow unmanned
aerial vehicles (UAVs) to navigate independently in
intricate surroundings.
Wang and Chen (2019) concentrate on employing deep
reinforcement learning techniques to make decisions for
autonomous drone navigation in unfamiliar surroundings.
The researchers investigate the use of artificial
intelligence methods to empower unmanned aerial
vehicles (UAVs) to make well-informed choices when
traversing unknown landscapes. They highlight the
significance of drawing insights from previous encounters
to enhance navigation capabilities. In their study, Kim and
Park (2020) explore the application of deep reinforcement
learning in path planning and obstacle avoidance for
unmanned aerial vehicles (UAVs). They emphasize the
capability of sophisticated learning algorithms to facilitate
UAVs in navigating through complex landscapes and
evading obstacles. The authors analyze the advancement
of deep learning methods in the context of effective path
planning and obstacle detection for unmanned aerial
vehicles (UAVs). They highlight the significance of
strong algorithms in guaranteeing the safety of UAV
operations. Garcia and Rodriguez (2021) conducted a
survey on the use of deep learning algorithms for
autonomous navigation of UAVs. Their study provides
valuable information on the most advanced methods and
how they are applied in different real-world situations.
The authors explore the application of deep neural
networks in tasks such as object detection, localization,
and mapping, emphasizing the potential of deep learning
in improving the navigation abilities of unmanned aerial
vehicles (UAVs). Zhang and Wang (2022) concentrate on
employing deep Q-networks to facilitate real-time
decision-making for drone navigation in dynamic
situations. The authors present an innovative method that
enables unmanned aerial vehicles (UAVs) to efficiently
and precisely analyze real-time data, facilitating their safe
navigation in dynamic surroundings. Lee and Lee (2023)
conducted a survey that explores the utilization of deep
learning techniques for autonomous navigation of UAVs.
The survey focuses on the newest breakthroughs in deep
learning and its applications in UAV navigation. The
researchers investigate a range of deep learning structures
and algorithms, emphasizing their advantages and
disadvantages in diverse navigation situations. They
emphasize the practical use of these algorithms in real-
world unmanned aerial vehicle (UAV) operations. In this
study, Wang and Zhang (2019) present a navigation
approach for unmanned aerial vehicles (UAVs) in
situations where GPS signals are absent or limited. The
proposed strategy utilizes reinforcement learning
techniques to overcome the obstacles associated with
navigation in such environments. The authors explore the
application of reinforcement learning algorithms in
facilitating unmanned aerial vehicles (UAVs) to travel
independently without depending on GPS signals. They
highlight the significance of resilient navigation strategies
in guaranteeing the accomplishment of missions. Li and
Liu (2020) propose a path planning algorithm for
numerous UAVs in urban contexts that utilizes swarm
Amit Joshi. et al. AI-Enabled Drone Autonomous Navigation and Decision Making For Defence Security
intelligence. The program specifically emphasizes
cooperative navigation strategies. The authors examine
the creation of swarm intelligence algorithms that draw
inspiration from the collective behaviour of natural
swarms. They emphasize the usefulness of these
algorithms in coordinating several UAVs for effective
navigation in urban situations. Zhou and Wu (2021)
investigate the utilization of fuzzy logic systems to make
decisions for unmanned aerial vehicles (UAVs) in
environments with uncertainty. The study highlights the
capacity of fuzzy logic to effectively handle uncertain
situations. The authors present a decision-making
framework based on fuzzy logic, which enables
unmanned aerial vehicles (UAVs) to make reliable
decisions in situations with high levels of uncertainty.
This framework takes into account several elements,
including weather conditions, sensor noise, and
communication delays. Kim and Lee (2022) examine the
use of learning-based techniques in UAV navigation,
where human involvement is incorporated to improve
safety. The authors emphasize the significance of
integrating human expertise into autonomous navigation
systems. The authors suggest a hybrid methodology that
integrates machine learning algorithms with human
intervention to enhance the safety and dependability of
unmanned aerial vehicle (UAV) navigation systems.
Zhang and Wang (2023) examine the use of multi-agent
reinforcement learning to facilitate cooperative navigation
of unmanned aerial vehicles (UAVs) in environments with
obstacles, focusing on the challenges of coordinating
many UAVs. The authors present a multi-agent
reinforcement learning framework that enables unmanned
aerial vehicles (UAVs) to learn collaboratively and
adaptively navigate across complex settings. The study
highlights the significance of cooperation and
coordination among UAVs. In their 2018 publication,
Huang and Li provide an overview of vision-based
autonomous navigation systems for unmanned aerial
vehicles (UAVs), with a particular focus on the
significance of visual sensors in facilitating self-governing
flying. The authors explore the advancement of vision-
based navigation algorithms for tasks such as identifying
obstacles, determining location, and creating maps. They
emphasize the practical uses of these algorithms in a range
of unmanned aerial vehicle (UAV) operations.
E. Study purpose
The purpose of the study is to develop and validate an
advanced AI-enabled drone system tailored for
autonomous navigation and decision making in defense
security contexts. In response to the evolving nature of
security threats and the increasing complexity of
operational environments, the study aims to leverage
cutting-edge technologies to enhance the capabilities and
effectiveness of defense security operations.The primary
objective is to design and implement a comprehensive
system that enables drones to autonomously navigate
through diverse terrains, detect potential threats, and make
informed decisions in real-time. By integrating advanced
AI algorithms, the study seeks to empower drones with
the ability to adapt to dynamic environments, respond
proactively to emerging threats, and execute missions
with precision and efficiency. The study aims to address
key challenges and requirements associated with the
deployment of autonomous drone systems in defense
141
security applications. This includes ensuring reliability,
safety, and compliance with ethical and legal standards
throughout the development and deployment lifecycle. By
incorporating fail-safe mechanisms, robust decision-
making frameworks, and adherence to regulatory
guidelines, the study seeks to build trust and confidence in
the capabilities and responsible use of autonomous drone
technologies.
Additionally, the study aims to demonstrate the
operational effectiveness and practical utility of the
developed AI-enabled drone system through extensive
simulation, testing, and field exercises. By showcasing the
system's performance in diverse scenarios and
environments, the study aims to validate its potential to
augment human capabilities, enhance situational
awareness, and improve mission outcomes in defense
security operations. The study endeavors to contribute to
the advancement of defense security capabilities through
the development and validation of innovative AI-enabled
drone technologies. By addressing critical operational
requirements and challenges, the study aims to pave the
way for the responsible integration and utilization of
autonomous systems in modern defense and security
strategies, ultimately enhancing national security and
safeguarding critical assets and interests.
II. MATERIALS AND METHODS
In conducting a review paper on AI-enabled drone
autonomous navigation and decision making for defence
security, a comprehensive analysis of existing literature
and research findings was carried out. The materials used
in this study include a wide range of academic articles,
research papers, conference proceedings, and reports
focusing on AI applications in drone navigation and
decision-making within defence contexts. Sources were
collected from reputable databases such as IEEE Xplore,
Scopus, and Google Scholar to ensure the inclusion of
high-quality, peer-reviewed publications. The review
process involved the examination of various aspects of
AI-enabled drone navigation and decision-making
systems. This included an in-depth analysis of different
AI techniques such as machine learning, deep learning
and reinforcement learning that have been applied to
autonomous drone navigation. Specific attention was
given to the components and functionalities of these
systems, including path planning, obstacle avoidance,
sensor fusion, and real-time decision-making in complex
environments. The review also explored the working
mechanisms of AI in drone navigation, such as vision-
based and sensor-based methods, data acquisition, and
processing, and the integration of AI algorithms for
efficient and reliable autonomous operations.
Furthermore, the study assessed the practical applications
of AI-enabled drone navigation in defence-security,
including reconnaissance, surveillance, and tactical
operations. Through a systematic review of the literature,
key trends, challenges, and future research directions
were identified. This involved a critical evaluation of the
current state of AI technologies in drone navigation and
decision making, as well as potential areas for
improvement and innovation in the defence security
sector. The review aims to provide a comprehensive
 Environment. Technology. Resources. Rezekne, Latvia
Proceedings of the 15th International Scientific and Practical Conference. Volume IV, 138-143
understanding of the advancements in AI-enabled drone
navigation and decision making and its implications for
defence security applications.
III. RESULTS AND DISCUSSION
The review paper on AI-enabled drone autonomous
navigation and decision-making for defence security
presents a detailed examination of various AI techniques,
their components, and their impact on drone navigation
and decision-making in defence contexts. The results of
the review showcase a broad range of AI methods applied
to drone navigation and decision-making, including
machine learning, deep learning, reinforcement learning,
and fuzzy logic systems. These methods have been
instrumental in enhancing the capabilities of drones for
defence operations, including reconnaissance,
surveillance, and tactical missions. One key finding from
the review is the significant progress made in vision-
based and sensor-based navigation systems, which allow
drones to perceive their environment and make real-time
decisions. The integration of AI algorithms with sensor
data has led to improved path planning, obstacle
avoidance, and target tracking capabilities. For instance,
deep learning-based computer vision techniques have
shown promise in enabling drones to identify and classify
objects in complex environments, thereby improving
their operational effectiveness. The review also highlights
the challenges associated with AI-enabled drone
navigation, such as the need for robust and reliable
algorithms that can operate in dynamic and uncertain
environments. These challenges include managing the
trade-offs between speed and accuracy in decision-
making, ensuring the safety and security of autonomous
operations, and addressing ethical and legal
considerations. Despite these challenges, the review
identifies several areas for future research and
development, including the advancement of multi-agent
systems for cooperative drone navigation and the
exploration of hybrid AI approaches that combine
different techniques for optimal performance.
Additionally, the integration of AI-enabled drones with
other defence systems, such as ground-based sensors and
communication networks, offers potential for enhanced
situational awareness and decision-making capabilities.
The review demonstrates the significant potential of AI-
enabled drone autonomous navigation and decision-
making for defence security applications. By providing
an overview of the current state of AI technologies and
their impact on drone navigation, the review offers
insights into emerging trends and opportunities for
further research and development in this field.
IV. CONCLUSION
The emergence of autonomous drone detection and
navigation systems is a notable progression in defence
technology, addressing the growing ubiquity of unmanned
aerial vehicles (UAVs) in defence, security, and
commercial domains. By harnessing artificial intelligence
(AI) technology, these technologies empower drones to
independently identify, monitor, and manoeuvre through
intricate surroundings. Artificial intelligence (AI)
142
algorithms, such as Deep Reinforcement Learning (DRL),
Convolutional Neural Networks (CNNs), Recurrent
Neural Networks (RNNs), Fuzzy Logic Systems,
Bayesian Networks, Evolutionary Algorithms, and Model
Predictive Control (MPC), are crucial in improving the
capabilities of drones. DRL algorithms such as Deep Q-
Networks (DQN) and Proximal Policy Optimization
(PPO) allow drones to acquire optimal navigation
strategies by repeatedly attempting different approaches,
enabling them to explore intricate terrains and make
instantaneous choices. Convolutional Neural Networks
(CNNs) interpret visual data captured by cameras on
board to detect and classify targets, obstacles, and
landmarks, hence aiding in navigation and decision-
making tasks. Recurrent Neural Networks (RNNs)
analyze sequential data, enabling drones to predict
alterations in the surroundings and adapt their navigation
tactics accordingly. Fuzzy logic systems are designed to
manage and account for uncertainty in the process of
making decisions. On the other hand, Bayesian networks
are used to represent and analyze the various aspects that
contribute to risk, enabling more informed decision-
making. Evolutionary algorithms are used to improve the
efficiency of route planning activities, while Model
Predictive Control (MPC) is employed to forecast the
future behaviour of a system for trajectory planning. By
integrating these algorithms, military drones are equipped
with improved situational awareness, adaptable navigation
capabilities, and intelligent decision-making in
challenging and ever-changing circumstances.
Autonomous drone detection and navigation systems play
a vital role in defence applications, ensuring the safety of
infrastructure, monitoring borders, and shielding military
assets from aerial threats. These technologies improve the
ability of defence personnel to be aware of their
surroundings, allowing them to quickly identify
unauthorized drones and respond appropriately. By
effectively manoeuvring through intricate airspace and
making immediate decisions using data, these systems
reduce security risks, guaranteeing the safety and
protection of military personnel and equipment.
Autonomous drone detection and navigation systems
employ AI-powered technologies such as computer
vision, sensor fusion, and decision-making algorithms.
These systems offer defence forces pre-emptive methods
to prevent airborne threats from hostile drones. These
technologies enhance the defensive capabilities of
governments and organizations by promptly identifying
possible hazards, precisely selecting targets, and
effectively coordinating responses. Autonomous drone
detection and navigation systems are a significant
development in defence technology.
V. REFERENCES
[1] J. D. Smith and R. W. Jones, "Autonomous Navigation for
Unmanned Aerial Vehicles: A Review of Recent Advances,"
*Journal of Intelligent Robotics*, vol. 45, no. 2, pp. 123-135,
2018.
[2] L. Wang and Q. Chen, "Deep Reinforcement Learning-Based
Decision Making for Autonomous Drone Navigation in Unknown
Environments," *IEEE Transactions on Intelligent Transportation
Systems*, vol. 21, no. 4, pp. 1678-1687, 2019.
[3] S. Kim and H. Park, "Path Planning and Obstacle Avoidance for
UAVs using Deep Reinforcement Learning," *International
Journal of Control, Automation, and Systems*, vol. 18, no. 5, pp.
1201-1212, 2020.
Amit Joshi. et al. AI-Enabled Drone Autonomous Navigation and Decision Making For Defence Security
[4] M. Garcia and A. Rodriguez, "Autonomous Navigation of UAVs
using Deep Learning Techniques: A Survey," *IEEE Transactions
on Aerospace and Electronic Systems*, vol. 57, no. 3, pp. 1021-
1035, 2021.
[5] Y. Zhang and C. Wang, "Real-time Decision Making for Drone
Navigation in Dynamic Environments using Deep Q-Network,"
*Robotics and Autonomous Systems*, vol. 144, pp. 102-115,
2022.
[6] K. J. Lee and H. Lee, "A Survey on Deep Learning-Based
Approaches for Autonomous Navigation of UAVs," *Journal of
Robotics and Autonomous Systems*, vol. 90, pp. 78-92, 2023.
[7] X. Chen and Q. Liu, "Path Planning for UAVs in Dynamic
Environments: A Review," *IEEE Transactions on Aerospace and
Electronic Systems*, vol. 54, no. 6, pp. 2458-2469, 2018.
[8] Z. Wang and L. Zhang, "Reinforcement Learning-Based
Navigation Strategy for UAVs in GPS-Denied Environments,"
*Journal of Navigation*, vol. 72, no. 3, pp. 567-580, 2019.
[9] W. Li and M. Liu, "Swarm Intelligence-Based Path Planning
Algorithm for Multiple UAVs in Urban Environments," *Applied
Soft Computing*, vol. 88, p. 106042, 2020.
[10] Y. Zhou and H. Wu, "Decision Making for UAVs in Uncertain
Environments using Fuzzy Logic Systems," *Expert Systems with
Applications*, vol. 176, p. 114831, 2021.
[11] S. Kim and J. Lee, "Learning-based UAV Navigation with Human
Intervention for Enhanced Safety," *IEEE Robotics and
Automation Letters*, vol. 7, no. 3, pp. 4947-4954, 2022.
[12] Q. Zhang and Y. Wang, "Multi-agent Reinforcement Learning for
Cooperative UAV Navigation in Cluttered Environments," *IEEE
Transactions on Cybernetics*, vol. 53, no. 1, pp. 120-132, 2023.
143
[13] X. Huang and S. Li, "Vision-Based Autonomous Navigation
System for UAVs: A Review," *Journal of Intelligent & Robotic
Systems*, vol. 90, no. 1, pp. 213-226, 2018.
[14] Y. Wu and J. Wang, "Autonomous Navigation of UAVs using
Cooperative Localization Techniques," *Sensors*, vol. 19, no. 5,
p. 1182, 2019.
[15] H. Zhang and K. Liu, "Decision Making for UAVs in Emergency
Situations: A Hybrid Fuzzy Logic and Deep Reinforcement
Learning Approach," *Journal of Intelligent & Fuzzy Systems*,
vol. 39, no. 2, pp. 1459-1471, 2020.
[16] J. Li and H. Chen, "Simultaneous Localization and Mapping for
UAV Navigation in GPS-Denied Environments: A Review,"
*IEEE Access*, vol. 9, pp. 102319-102335, 2021.
[17] X. Wang and M. Zhang, "Model Predictive Control for
Autonomous UAV Navigation in Complex Environments,"
*Control Engineering Practice*, vol. 122, p. 104933, 2022.
[18] Y.Liu and S.Yang,"Evolutionary Algorithm-Based Path Planning
for UAVs in Dynamic Environments," *Engineering Applications
of Artificial Intelligence*, vol.104, p.104309, 2023
[19] J. Park and H. Kim, "Learning-Based Decision Making for
Autonomous UAV Navigation in Urban Environments," *Journal
of Field Robotics*,vol. 35, no. 5, pp. 784-798, 2018.
[20] L. Zhao and B. Wang, "Dynamic Path Planning for UAVs in
Unknown Environments using Bayesian Networks," *IEEE
Transactions on Systems, Man, and Cybernetics: Systems*, vol.
49, no. 11, pp. 2474-2486, 2019.