Alghamdi_Munir_UAV_Architecture_av_IEEE_CEM_2021
Alghamdi_Munir_UAV_Architecture_av_IEEE_CEM_2021
Alghamdi_Munir_UAV_Architecture_av_IEEE_CEM_2021
net/publication/349826625
CITATIONS READS
51 2,019
3 authors:
Hung Manh La
University of Nevada, Reno
200 PUBLICATIONS 6,035 CITATIONS
SEE PROFILE
All content following this page was uploaded by Arslan Munir on 28 June 2021.
Abstract—Unmanned aerial vehicles (UAV) have been in more critical applications, such as search and rescue
gaining significant attention in recent times as they are missions, and delivering medical supplies. UAVs have
becoming increasingly accessible and easier to use. UAVs been used in military applications since the early 1970s,
are used in a variety of applications, ranging from civilian and their role is growing now more than ever. As of 2018,
tasks, law enforcement, and rescue applications, to mili-
the number of remotely piloted aircraft in the U.S. Air
tary reconnaissance and air strike missions. This article
serves as an introduction to UAV systems’ architecture,
Force rose from 1,366 in 2013 to 2,404 [2].
classification, and applications to help researchers and UAVs have been the subject of considerable research
practitioners starting in this field get adequate information over time and various UAV topics have been covered.
to understand the current state of UAV technologies. The Shakhatreh et al. [3] research covered in-depth the
article starts by inspecting the UAVs’ body configuration civilian applications of UAVs and the key challenges
styles and explains the physical components and sensors and problems that are facing each application. The
that are necessary to operate and fly a UAV system. The authors provided a classification of UAVs based on
article also provides a comparison of several components the communication platform into low altitude platform
for state-of-the-art UAVs. The article further discusses
(LAP) and high altitude platform (HAP). Kim et al. [4]
different propulsion methods and various payloads that
could be mounted on the UAV. The article then explores the provided a comprehensive study on the use of UAVs in
classification of UAVs followed by the application of UAVs agriculture. They defined a basic UAV architecture based
in different domains, such as recreational, commercial, on body type as fixed-wing or rotatory wing. The authors
and military. Finally, the article provides a discussion of explored the hardware, sensors, and communication
futuristic technologies and applications of UAVs along with systems that are used in agricultural UAVs. The authors
their associated challenges.1 further discussed the limitations of UAVs in agricultural
applications. Gupte et al. [5] surveyed quadcopter UAVs.
The authors discussed the flying mechanism, control
I. I NTRODUCTION
systems, and some of the sensors of quadcopter UAVs.
commercial, and military domains. installed behind the UAV. Another style is a delta wing,
Our main contributions in this article are as follows: which is the simplest build configuration. This style has
• Discussion of UAV architecture and components the propeller at the back like the pusher configuration
including UAV body styles, sensors, flight but there is no rudder and the elevons control the pitch
controller, propulsion methods, and payload. and yaw.
• Providing a comparison of several components for Multirotor: Multirotor UAVs (Fig. 1) can be configured
state-of-the-art UAVs. using multiple motors installed on arms. This body type
• Examining different classifications of UAVs such has the advantages of vertical takeoff and landing and
as the one based on DOD group classification and can hover in a fixed position. Multirotor UAVs can fly
the others based on UAV weight, wingspan, and in any direction and change speed, altitude, and direction
altitude. abruptly, but these UAVs usually tend to have a shorter
• Elaboration of a variety of recreational, commercial, range and flight duration. The multirotor body can be
and military applications of UAVs. designed as a traditional helicopter, a tricopter where
• Discussion of futuristic embodiments and it has three arms with three motors one on each, or a
applications of UAVs and the challenges associated quadcopter where it has four arms with four motors.
with these technologies. Multirotor UAVs can also be configured with a variable
number of arms and rotors.
II. UAV A RCHITECTURE Vertical Takeoff and Landing: Vertical takeoff and
Most UAVs are made up of modular components that landing (VTOL) body type (Fig. 1) is very versatile
fit on and inside the body of UAVs and are powered as it can be a hybrid of the two previously mentioned
by an energy supply. Depending on the style of flight body styles (i.e., fixed-wing and multirotor). The VTOL
and missions to be performed, a UAV can have one UAVs typically use multirotors to takeoff, land, and
of many different body configurations. This section hover, whereas these UAVs transition to a normal fixed-
discusses UAV body styles, body material, and different wing flight style to travel horizontally. The VTOL UAVs
components of UAVs. can be configured by affixing quadcopter rotors on a
fixed-wing like Aerosonde HQ by Textron Systems.
A. UAV Body Styles Other VTOL body configurations include tiltrotors and
tiltwings. In tiltrotor UAVs, some of the rotors on the
UAVs can have many different body styles or shapes
wing tilt vertically and horizontally like the Bell V-247
depending on the flight style and intended usage or
Vigilant. Tiltwing UAVs have a wing that is normally
missions. UAV body styles mainly fall into one of the
horizontal while flying but rotates up in the cases of
following types:
vertical takeoff and landing.
Easy to
Foam 0.05 46 – 60 shape, Weak
Flexible
Carbon
1.75 2000 – 5600 Very
Fibers measure the UAV’s speed, orientation, and direction
lightweight, Expensive
G-10 1.80 262 – 310 Strong along 9 axes. Currently, most professional and military
IMUs tend to incorporate redundant IMU sensors. The
Aluminum 2.70 276 – 310 Strong Heavy
advantage of this redundancy is to get the most accurate
data from the sensors. Furthermore, redundancy is
of these components are essential for the operation helpful in case of a sensor failure, where the redundant
and flying of UAVs whereas other components are components provide a level of fault tolerance.
installed for mission-specific purposes. Fig. 2 depicts
Global Positioning System: The global positioning
the main components of a UAV. This section discusses
system (GPS) chip is used to get the position of the
the main components of UAVs focusing on sensors,
UAV from multiple satellites, which is how the UAV
flight controllers, propulsion subsystem, payloads, and
gets a navigation system. The GPS chip together with
communication subsystem.
the IMU constitutes the navigation system of a UAV,
1. S ENSORS : UAVs use a variety of sensors that provide which measures the speed, positioning, and heading of
the flight controller with data such as telemetry, attitude the UAV. The GPS chip can either be a separate receiver
measurements, terrain, and obstacles. Sensors are also chip or part of the flight controller.
an integral part of situational awareness of UAVs. The Visual, Ultrasonic, and Infrared Sensors: UAVs often
combination of sensors installed on UAVs depends on the incorporate visual, ultrasonic, and infrared sensors to
complexity and sophistication level of UAVs. The most detect obstacles and avoid collisions by measuring the
common sensor suite for UAVs include the following relative distances to the objects in the path of the UAV
sensors. or surrounding it.
Inertial Measurement Unit: An inertial measurement
2. F LIGHT C ONTROLLER : The flight controller is the
unit (IMU) is an electronic device that provides a
brain of the UAV. The flight controller is the hardware
measurement of UAV’s attitude, specific force (i.e., non-
and the firmware that processes all the information that
gravitational force per unit mass also called g-force),
is collected from the IMU, GPS, and other sensors. The
angular velocity, orientation, and position information.
flight controller also controls the payload on the UAV
Most of the IMUs consist of a combination of the
(such as cameras and other sensors). The flight controller
following sensors:
can be fully autonomous or just provide some level of
• Accelerometer: measures the linear acceleration ax , navigation assistance in the autopilot mode using the
ay , az on the three axes of the body x,y,z. sensed data from the IMUs and other sensors to control
• Gyroscope: measures the angular change rate p, q, the UAV. Some of the flight controllers are capable of
r on the three axes of the body x,y,z. being programmed to set a home point from where the
• Magnetometer: Determines the heading of a UAV UAVs can be launched and land back automatically. The
by measuring the earth’s magnetic field. flight controllers provide the ability to fly a UAV through
• Barometer: Determines the altitude of a UAV by preloaded flight paths utilizing waypoints. Waypoints are
measuring the earth’s atmospheric pressure. a set of predefined location coordinates, which can be set
Since each one of the IMU components works on 3 up before the flight by specifying each point’s latitude,
axes, by combining the three components (accelerometer, longitude, and altitude, that creates a route so that a
gyroscope, magnetometer) the flight controller can UAV can fly autonomously from one waypoint to the
4
next. The flight controller can also be programmed to TABLE III: Different types of fuel-powered UAV
make the UAV loiter in a specific location. The flight engines.
controller automates the UAV flight with the help of Attribute LYCOMING Rotax 914 Honeywell Rolls
sensors, throttle, servos, and payloads that are connected EL-005 T76 Royce
to the flight controller. The autopilot capabilities depend F137
on the flight controller’s firmware. All the commands Two-stroke Four-stroke Turboprop Turbofan
are programmed on a computer before the flight and Type piston piston jet jet
then loaded to the flight controller. Table II shows the engine engine engine engine
processing power, memory, IMU components, and cost Max 900 Watts 84.5 kW @ 700 kW 44.7 kW
of three of the most common flight controllers: the Power @ 5,500 5,800 RPM @ 2,000 @ 16300
Pixhawk Cube Orange, Lockheed Martin Kestrel v3.1, RPM RPM RPM
and CUAV V5 NANO. Weight 13.8 lb 140.8 lb 385 lb 641 lb
TABLE IV: U.S. DOD Classification of UAV Groups. TABLE V: UAV classification based on weight and
wingspan [9].
UAV Group Max Takeoff Operating Speed
Weight (lb.) Altitude (Ft) (knots) Class Weight WingSpan
Group 1 < 20 < 1200 < 100 UAV > 5Kg > 2m
Group 2 21-55 < 3500 Small UAV ≤ 5Kg ≤ 2m
< 250
Group 3 < 1320 MAV (Micro Air Vehicle) ≤ 2Kg ≤ 1m
< 18,000
Group 4 Any NAV (Nano Air Vehicle) ≤ 50g ≤ 15m
> 1320
Group 5 > 18,000 Speed PAV (Pico Air Vehicle) ≤ 3g ≤ 2.5cm
Smart Dust ≤ 0.5g ≤ 0.25cm
TABLE VII: Comparisons of active military UAVs. could be semi-autonomous or fully-autonomous. Semi-
Attribute RQ-4 MQ-1C MQ-1 ScanEagle
autonomous swarms are usually controlled by a GCS
Global Gray Predator which communicates with each UAV simultaneously.
Hawk Eagle Fully-autonomous swarms use flying ad-hoc networks
Max Speed 340 167 120 80 (FANET) to communicate with each other and one leader
(knots) UAV communicates with the GCS. Utilizing swarms in
Max Altitude 60,000 25,000 25,000 19,500 applications such as surveying and mapping, agricultural,
(ft) and logistics and delivery has the benefits of saving
Engine Rolls Lycoming Rotax 2-stroke time and man-hours by distributing the tasks between
Royce DEL-120 914 3W the UAVs. UAV swarms have gained a lot of attention
F137-RR- piston
from the Air Force and the Department of Defense in
100
recent years. UAV swarms can even be used to save lives
Max Internal 3,000 540 450 7.5
Payload (lb.)
in the case of search and rescue applications. Deploying
a swarm of UAVs in a disaster zone leads to a faster
Max NA 1,500 300 NA
External discovery of survivors. The main challenges in UAV
Payload (lb.) swarms are collision avoidance between the UAVs in the
Endurance 32+ 50+ 24 24+ swarms and the energy efficiency of swarm operation.
(Hours)
Cost $130M $21.5M $40M $3.2M B. Augmented Reality
C. Military AR is a technology that allows users to interact with
computer-generated objects in a real-world environment
A UAV is considered a military type if it is [10]. AR has been used in a variety of applications,
flown or manufactured for military purposes, such such as design and manufacturing, tourism, navigation,
as intelligence gathering, surveillance, reconnaissance, education, and gaming. By integrating AR technology in
target acquisition, and battle damage assessment, or if UAVs, a multitude of new applications are possible. AR
the UAV can carry a variety of weapons and perform can be used to display real-time overlaying information
air strikes. Some of the types of military UAVs that on the real-world environment like maps, floor plans,
are being used right now are the General Atomics MQ- and points of interest. These AR maps can be beneficial
1 Predator, Northrop Grumman RQ-4 Global Hawk, in the military, and search and rescue operations.
General Atomics MQ-1C Gray Eagle, and Boeing Insitu Another example of integrated UAV-AR is planning and
ScanEagle. Table VII shows a comparison of the four designing, where architects and engineers can fly over
above mentioned UAVs in terms of maximum speed, a location and use 3D models to visualize the project.
maximum altitude, engine type, maximum internal and Integrated UAV-AR can also be used in inspection where
external payloads, endurance, and cost. the AR would overlay and synchronize the building
information model (BIM) on the video feed from the
V. F UTURE OF UAV S AND A SSOCIATED UAV. This could be useful in inspecting underground
C HALLENGES pipes, bridges, power lines, and dams.
In future, UAVs will be integrated with emerging
technologies such as augmented reality (AR), artificial C. Artificial Intelligence
intelligence (AI), and cellular networking to enable UAVs using AI can be utilized in various object
a multitude of innovative applications. Furthermore, detection and recognition applications. Detecting traffic
advancements and miniaturization of electronics will congestion is an example of an application utilizing
help realize exciting new applications of UAVs. This AI-assisted UAVs [11]. In this example, a UAV
section discusses some of the futuristic embodiments and takes pictures of traffic in a preplanned route, which
applications of UAVs and the challenges associated with are subsequently processed by a convolutional neural
them. network (CNN) to decide if a road is congested or not.
The results from the CNN are then sent to the road traffic
management center, which can take necessary actions to
A. UAV Swarms alleviate the congestion. Another application of UAVs
A swarm of UAVs is a group of UAVs that coordinate utilizing AI is early forest fire recognition [12]. In this
with each other to carry out a task. The swarm system, a UAV or a group of UAVs use cameras to detect
9
fires. The AI in UAVs is trained to recognize smoke unknown area, while simultaneously updating the map
and fire in pictures. The resulting alerts are then sent and tracking the current location of a vehicle within the
to the GCS. The integration of AI in UAVs has many created map. SLAM is an important topic in unmanned
challenges, the biggest of which is computational power vehicle systems as it is one of the main methods used in
as UAVs have limited on-board computation ability creating autonomous vehicles. SLAM is also important
as well as limited battery energy for battery-operated for navigating UAVs in GPS-deprived locations or
UAVs. Due to the limited on-board compute capability, flying UAVs indoor. To obtain sensor data required for
AI embedded in UAVs is often not very sophisticated developing SLAM for UAVs, a UAV could be outfitted
and thus can have a limited accuracy. An incorrect with many sensors, such a LiDAR sensor, an RGB-D
prediction/classification could result in wrong alerts and camera, a stereo camera, or any combination of these
wasted efforts in case of false positives and no response sensors. The SLAM algorithm for autonomous UAVs
for an emergency situation in case of false negatives. can be based on Kalman filters or particle filters. The
algorithm uses the sensor data to determine the odometry
D. Wireless Networking and update the map. High flight speed and real-time
response are challenges that currently face SLAM in
UAVs are projected to be a major part of the 5G and UAVs.
beyond 5G (B5G) cellular architectures [13]. UAVs can
connect to cellular networks which help improve the VI. C ONCLUSION
real-time data stream and extend the maximum range
With the price-to-value of building or buying an
of UAVs. Swarms of UAVs could utilize the existing 5G
unmanned aerial vehicle (UAV) being lower than ever,
and B5G networks to enhance communication between
and with the advancements in flight controllers and
the UAVs which leads to better autonomy. Another
payloads, it is now more feasible, even advantageous,
application is using a swarm of UAVs as flying base
to use UAVs for commercial and military missions. This
stations to provide cellular networks in an area deprived
article explored the architecture and components of a
of cellular signal or in disaster zones. Communication
modern UAV. The article presented some of the common
reliability, security, privacy, quality of service, and
flight controllers and engine types for contemporary
interference management are some of the challenges in
UAVs. The article also discussed different body styles
wireless networking using UAVs.
and body material types for UAVs. The article further
presented different classification methods for UAVs.
E. Nano-UAVs Finally, the article elaborated the applications of UAVs in
Highly miniaturized UAVs could be developed and recreational, commercial, and military domains as well
deployed either solo or in swarms and could be used as discussed the futuristic embodiments and applications
in military applications such as espionage or precision of UAVs along with their associated challenges.
strikes. For example, the FLIR Black Hornet PRS is
a nano-UAV with low visual and audible signatures, ACKNOWLEDGEMENTS
a size of only 13 grams, and is equipped with the This work is supported in part by the U.S. National
same cameras and sensors as a full-sized UAV. The Science Foundation (NSF) under grants NSF-CAREER:
Black Hornet’s small size allows it to carry surveillance 1846513 and NSF-PFI-TT: 1919127, and the U.S.
and reconnaissance missions covertly and unnoticed. Department of Transportation, Office of the Assistant
Another application for nano-UAVs is covert lethal force Secretary for Research and Technology (USDOT/OST-
missions and assassinations by equipping a nano-UAV R) under Grant No. 69A3551747126 through INSPIRE
with a small amount of explosives and targeting vital University Transportation Center.
organs. The Swiss Drones and Robotics Center tested
the possibility of using 3 grams of explosive charge R EFERENCES
and a detonator to assassinate a person by landing and [1] “FAA Aerospace Forecast Fiscal Years 2020-2040,” Federal
detonating on the head; the resulting injuries were fatal Aviation Administration, United States, Tech. Rep., 2019,
with no chances of survival. https://www.faa.gov/data research/aviation/aerospace
forecasts/media/Unmanned Aircraft Systems.pdf.
[2] “Unmanned Aerial Systems: Air Force Pilot Promotions Rates
F. UAV SLAM Have Increased but Oversight Process of Some Positions Could
Be Enhanced,” Federal Aviation Administration, Washington,
Simultaneous localization and mapping (SLAM) is D.C.:U.S. Government Printing Offices, Tech. Rep. GAO Pub-
a computational problem of creating a map of an lication No. 19-155, 2013.
10
[3] H. Shakhatreh, A. H. Sawalmeh, A. Al-Fuqaha, Z. Dou, E. Al- Magazine, vol. 8, no. 3, pp. 81–86, 2019.
maita, I. Khalil, N. S. Othman, A. Khreishah, and M. Guizani, [12] D. Kinaneva, G. Hristov, J. Raychev, and P. Zahariev, “Applica-
“Unmanned Aerial Vehicles (UAVs): A Survey on Civil Appli- tion of artificial intelligence in uav platforms for early forest fire
cations and Key Research challenges,” Ieee Access, vol. 7, pp. detection,” in 2019 27th National Conference with International
48 572–48 634, 2019. Participation (TELECOM). IEEE, 2019, pp. 50–53.
[4] J. Kim, S. Kim, C. Ju, and H. I. Son, “Unmanned Aerial [13] M. Mozaffari, A. T. Z. Kasgari, W. Saad, M. Bennis, and
Vehicles in Agriculture: A Review of Perspective of Platform, M. Debbah, “Beyond 5g with uavs: Foundations of a 3d wireless
Control, and Applications,” IEEE Access, vol. 7, pp. 105 100– cellular network,” IEEE Transactions on Wireless Communica-
105 115, 2019. tions, vol. 18, no. 1, pp. 357–372, 2018.
[5] S. Gupte, P. I. T. Mohandas, and J. M. Conrad, “A Survey of
Quadrotor Unmanned Aerial Vehicles,” in 2012 Proceedings of Yousef Alghamdi is a Ph.D. student in the Department
IEEE Southeastcon. IEEE, 2012, pp. 1–6.
[6] T. McGeer and J. Vagners, “Wide-scale Use of Long-range
of Computer Science at Kansas State University, Man-
Miniature AEROSONDEs Over the World’s Oceans,” The Insitu hattan, KS. Alghamdi has an M.S. degree in information
Group, Bingen, 1999. technology from North Carolina Agricultural and Tech-
[7] A. C. Watts, V. G. Ambrosia, and E. A. Hinkley, “Unmanned nical University and a B.S. degree in information tech-
Aircraft Systems in Remote Sensing and Scientific Research:
Classification and Considerations of use,” Remote Sensing, nology from King Abdulaziz University, Saudi Arabia.
vol. 4, no. 6, pp. 1671–1692, 2012. Contact him at: usef@ksu.edu
[8] U.S. Army, “Eyes of The Army: U. S. Army Roadmap for
Unmanned Aircraft Systems 2010-2035,” U.S. Army, United Arslan Munir is an Associate professor in the Depart-
States, Tech. Rep., 2010, https://fas.org/irp/program/collect/ ment of Computer Science at Kansas State University,
uas-army.pdf.
[9] M. Hassanalian and A. Abdelkefi, “Classifications, Applica-
Manhattan, KS. Munir has a PhD in electrical and
tions, and Design Challenges of Drones: A Review,” Progress computer engineering from the University of Florida,
in Aerospace Sciences, vol. 91, pp. 99–131, 2017. Gainesville. He is a senior member of IEEE. Contact
[10] N. Shabani, A. Munir, and A. Hassan, “Revolutionizing e- him at: amunir@ksu.edu
Marketing via Augmented Reality: A Case Study in Tourism
and Hospitality Industry,” IEEE Potentials, vol. 38, no. 1, pp.
Hung M. La is an Associate professor in the Department
43–47, January 2019.
[11] L. Jian, Z. Li, X. Yang, W. Wu, A. Ahmad, and G. Jeon, “Com- of Computer Science and Engineering at University of
bining unmanned aerial vehicles with artificial-intelligence tech- Nevada, Reno. La has a PhD in electrical and computer
nology for traffic-congestion recognition: electronic eyes in engineering from Oklahoma State University. He is a
the skies to spot clogged roads,” IEEE Consumer Electronics senior member of IEEE. Contact him at: hla@unr.edu