4/26/22, 10:07 PM Turnitin

Turnitin
Originality
Report
Processed on: 26-Apr-2022 22:04 EET   Similarity by Source
ID: 1821191474 Similarity Index
Internet Sources: 99%
Word Count: 3964
Submitted: 1 100% Publications:
Student Papers:
5%
100%

new By HASSAN HAYDER

ABDULHUSSIEN

100% match (student papers from 26-Apr-2022)

Class: Offshore Vessel's DP (MT482T) ALEXANDRIA (Moodle TT)
Assignment: Sameh Tawfik Abd Elfattah Mahmoud Part 1 (Moodle TT)
Paper ID: 1821185571

APPLICATION OF DRONES IN MARITIME INDUSTRY (FIRE FIGHTING) 1

ABSTRACT Fire is a major hazard to life, be it at-sea or ashore. In order to
tackle the fire induced situations and ensure the safety of the crew on-
board as well as on-shore, the following propositions are encompassed in
this paper: Dynamic Firefighting Drone (Octocopter): A drone used for
autonomous firefighting applications on Manned and Unmanned Ships, and
even on ports, using Artificial Intelligence System (AIS) and IoT (Internet
of Things). The fire is tackled by automatically releasing DCP (Dry
Chemical Powder) or by the Wave Extinguisher System (consisting of Sonic
waves), in accordance to the situation presented and in conjunction with
the AIS. On that account, it can autonomously fight or limit the fire, till a
crew member arrives at the sight. Furthermore, it works as a guidance
mechanism for seafarers in close vicinity of the fire or at risk in general
and allows them to follow the drone to the nearest safe exit, avoiding
obstacles on the selected path. Artificial Intelligence System: The entire
set up utilizes Artificial Intelligence in the Octocopter and furthers the
development through Neural Network Learning. Smart Rescue Equipment:
The equipment works in tandem with the Drone and AIS. Vitality Band is
the proposed arm band with a vitals sensor, a fall sensor and a positioning
sensor, used in order to send immediate help to crew transponder attached
in clothing or shoes which can be used to send automated distress signals
to rest of the crew for immediate help. 2 KEYWORDS Drones, Artificial
Intelligence, Fire Fighting Technology, Internet of Technology, Automated
Firefighting Equipment, Smart Rescue Equipment, Neural Net 3
INTRODUCTION Fire has always been an unwelcome visitor onboard ship.
Despite the ever-increasing norms and precautions on fire-related
accidents, the accidents do occur and are responsible for the loss of
billions of dollars’ worth of cargo and many lives at sea [2]. Since the
process of fighting the fire onboard ship is manual, the response time as
well the probability of the fire to acquire a dangerous intensity is
increased. Therefore, to be better equipped to fight a fire, Artificial
Intelligence can be used in tandem with the Internet of Things to make the
process automated. In the event of a fire, the AI will be the first one to
detect it and take necessary action, thus decreasing the role and the
subsequent damage to a human during fire-fighting. members using IoT.
Clothing/Shoe Transponder is also proposed, which is a certain
temperature, it forms carbon-fiber- reinforced polymer which has a very
high strength-to-weight ratio, and is extremely rigid. - Frame: The least
possible frame diameter to lift the necessary payload weight would be

https://api.turnitin.com/newreport_printview.asp?eq=0&eb=0&esm=0&oid=1821191474&sid=0&n=0&m=2&svr=51&r=50.48832845325519&lang=… 1/7
4/26/22, 10:07 PM Turnitin

1250 mm. For optimum utility these are paired with 165 mm Rail Mount
and foldable legs. The said frame is made from Carbon Fiber. - Motor: An
Octocopter requires eight motors to fly. The drone uses brushless motors
since they are lighter on the - Drones Figure 1: Comparison of Materials
for the body of the Drone (Source: MDPI Journal) A drone is an aircraft of
the unmanned variety and is made of light composite materials to reduce
the operational weight and increase maneuverability. The said drones are
equipped with different state of the art technology such as infra-red
sensors, Self-Learning Artificial Intelligence, 3-D Mapping Technology and
location transponders to accommodate for dynamic rolling and
pitching[1].The drones only work when automatic alarms are activated
and are otherwise placed on charging stations and hence cause negligible
distraction to the crew. Due to flight accessibility the drone works
anywhere in the engine room including staircases, enclosed spaces and
places which aren't easily accessible. - Construction [3] The chassis of the
drone needs to be made of a lightweight, heat resistant and sturdy
material while also being economical. With an inclusion of metal
compounds, thermosetting plastics, the most viable recommendation is
Carbon Fiber Composites. Carbon fiber composites are comparatively heat
resistant, lightweight while being resilient to physical trauma. Carbon
fibers are usually combined with other materials to form a composite.
When infused with a plastic resin and baked at a battery.
Recommendation: Tarot 5008/340kv motors as they are economical and
reliable - ESCs or electronic speed control: These are essential pieces of
your drone as they are in charge of delivering power to the motors.
Recommendation: Hobby wing X Rotor 40A- OPTO. - Propellers: In order
to allow flight to the drone, propellers are used to provide aerodynamic
ability. Recommendations: Arms Diameter: 25mm, Motor to motor
diameter: 1200m, Height: 380cm, Body size: 250 x 240mm. - Connectors:
You will need 3.5 mm connectors to weld the motors and ESCs, and 4.5
mm connectors for the power distribution board and the power distribution
board. - Batteries: Recommendation: 2 x 22.4V x12000mAh Li-ion
batteries. One battery is primary whereas the other one is auxiliary. These
are also capable of being wirelessly charged. These batteries provide 40
minutes (with allowance of 5 minutes) of flying time with all the
functionalities. Figure 2: Diagrammatical Representation of a Radio - Gyro
System: The gyroscope system should work almost instantly to the forces
moving against the drone (gravity, wind etc.) to keep it stabilized. Apart
from that, the system provides navigational information to the flight
control systems. Recommendation: DJI A2 multi-rotors Thermal Imaging
Camera: A Thermal Imaging Camera will aid in better identification of fire,
humans and machinery at risk of catching fire. Recommendation: Seek
Compact Thermal Imager Autopilot System: This will be required to aid the
AI during flight time. Recommendation: PixHawk Flight Controller System.
4 ACTIVATION The ship's pre-existing fire alarm system is used to activate
the drone into action. Whenever a fire in a particular section of the engine
room is ignited, it is detected by the appropriate fire detecting system as
per SOLAS, which might vary from ship to ship {4}. These include: a.
Flame detectors: Detects flicker frequency of flame (25 Hz) b. Heat
Detectors: Uses bi-metallic type detecting elements c. Smoke Detectors:
1. Light obstruction type, 2. Ionization type. {S}) This information is fed
through pre- existing system for transferring it to the engine control room
i.e., electrical signals. Using an Ultra High Frequency Emitter (ranging
between 300MHz to3GHz) located near the drone base; the information
can be wirelessly transmitted to the drones which act as the receivers. The
Artificial Intelligence System then decides how the drone should engage
the fire. Figure 3 : Diagrammatical Representation of a Neural Net
(Credits: www.astroml.org) which will be a radially transmitted signal. At
its very core, Artificial Intelligence refers to giving decision- making
capabilities to a machine to give it autonomy. This AI system focuses on
the specific goal of Natural Language Processing. It alters a machine in

https://api.turnitin.com/newreport_printview.asp?eq=0&eb=0&esm=0&oid=1821191474&sid=0&n=0&m=2&svr=51&r=50.48832845325519&lang=… 2/7
4/26/22, 10:07 PM Turnitin

terms of perception and reasoning and develops intelligence. 4.1 Working

In reference with Figure 3, the AIS first establishes the central concept of
the initial signal. It interacts with the user to resolve any information gaps,
contradictions and ambiguities. It does so by providing inquiries which are
to be answered by the user in the form of additional information. This
additional information is added to the initial statement. Later, the AIS may
provide responses which are in accordance to the request made and are
based upon the understanding of the meaning of the input statements
which are derived through constantly providing inquires. As the initial
input statement becomes more understood, the response time reduces.
Figure 4: A Flowchart depicting the working of AIS Artificial Intelligence
System (AIS)[S] It's a system that accepts an invocation for
information/action and provides an appropriate response/action, which is
based upon understanding of the request, Artificial Neutral Net (Self-
learning) [6,7,8] The AIS heavily relies on Neural Networks in order to
Self-Learn and adapt to situations through prior experience. It's a system
which is inspired by the biological neural networks; example - brains of
animals. The neural net will help the AIS "learn" to perform tasks by
taking into consideration day- to-day examples. For example, in image
recognition, it might learn to identify images that contain a certain type of
fire by analyzing example images that have been manually labeled as
"electrical fire" or "no electrical friending the results to identify fires in
other images, without having prior knowledge about fires. Types of Neural
Networks suitable for onboard applications Convolutional Neural Network:
It is a deep, feed-forward networks, composed of one (or more)
convolutional layers where the layers are fully connected. The architecture
which comprises of tied weights and pooling layers allows CNNs to take
advantage of the 2D structure of the data. CNNs show superior results in
applications such as image and speech. These can be trained with
standard backpropagation (a method used to calculate a gradient that is
needed in the calculation of the weights to be used in the network). CNNs
can be trained easily than other feed-forward neural networks and
estimation of parameters is easier. Deep- Dreamland Robot Navigation are
one of the many applications of the same. Deep Belief Net: A deep belief
network (DBN) is a generative model made up of multiple layers of hidden
units. A DBN can be used to generatively pre-train an ANN by using the
learned DBN weights as the initial DNN weights. This is particularly helpful
when training data are limited, because poorly initialized weights can
significantly hinder model performance. 5 PROJECTION MAPPING A pre-
requisite for the AIS to function is a projection map of the ship which will
help the drone to perceive obstacles and the ship environment better. The
complexity within the ship can be easily programmed to its spatial
dimensioning and thus, allow greater flexibility for the drone in aspect of
navigating and traversing. It’ll help the AIS to calculate the shortest
possible route for reaching the desired location and also help in rescue
operation of the crew, if need be. Blockchain Technology: Figure 5:
Depiction of the Block chain Technology (Source: What is Block chain
Technology? a CBINSIGHTS newsletter) "The block chain is an
incorruptible digital ledger of economic transactions that can be
programmed to record, not just financial transactions but virtually
everything of value." It’s like an internet with robustness, built- in. The
AIS can learn about numerous numbers of situations by simply accessing
the blocks in which this technology stores information. At the same time,
this information is also available to the AIS working on ships across the
globe and any new update made to the information by one, is read-able by
the other. The system lives in a state of consensus and checks itself after a
given interval of time. It cannot be corrupted by altering any unit of
information, despite being, by definition, "public". Movement The Drone is
an Octocopter which is a VTOL ("Vertical Take-Off and Landing") type. In
order to move, the drone is embedded with a drone vision system which
uses obstacle detection sensors to scan the surroundings, while software

https://api.turnitin.com/newreport_printview.asp?eq=0&eb=0&esm=0&oid=1821191474&sid=0&n=0&m=2&svr=51&r=50.48832845325519&lang=… 3/7
4/26/22, 10:07 PM Turnitin

algorithms and SLAM technology produce the images into 3D maps

allowing the flight controller to sense and avoid the object. Figure 6:
Working of RFID tags Obstacle Detection and Avoidance Technology RFID
Tags{9, 10}:To accommodate for the constantly moving environment,
additional Radio Frequency Identification emitters are installed onboard
that operate at very vast band of frequencies (433Hz - 2.45GHz) with a
readability range of 2.25 meters. These tags, being the size of a chocolate
bar, can be glued to the walls easily. The emitters can withstand a
temperature of upto 200oC and can absorb vibrational shocks as well.
When the ship rolls or pitches, the inside environment in which the drone
flies constantly change. The numerous RFID tags form a Cartesian system,
that accordingly changes and the AI of the shift in the pre-programmed 3-
D Map. Others: Systems are fusing one or more of the following sensors to
sense and avoid obstacles: a. Vision Sensor: The Vision Sensor works by
perceiving oncoming objects through the camera feed. b. LiDAR:The LiDAR
measures obstacle distance to target by illuminating it with multiple laser
pulses and measuring the reflected ones.j11] Real-time operations When
the fire alarm goes off in a particular section, a drone is automatically
deployed to the location. Upon reaching the zone, the thermal imaging
camera activates for better accuracy and assesses the degree of the fire.
Thermal Imaging works on the principle of measuring infra-red radiation
that radiates off a body. When the number of quanta is larger, the body
under observation is hotter. This perception of hotness aids in locating
through temperature. The AIS checks for the required number of
additional drones to deal with the fire and transmits the request for the
same. The thermal camera and other systems transmit a live video feed
through a LAN Wi-Fi Network with repeaters, placed strategically to create
a continuous and fast network, to the control rooms (Engine Control Room
and Bridge). Initial firefighting measures, depending on the type of fire
and its degree - DCP Release or Sonic Extinguisher -are immediately
deployed. In case of any Humans trapped in the area, the Drone clears a
pathway by extinguishing fire and leading them to the safest possible exit.
While fighting fire, the Drone's primary purpose is to completely extinguish
the flames.j13] 6 FIRE FIGHTING Dry Chemical Powder: Dry Chemical is a
powder composed of very small particles. The current system uses ABE
Dry Chemical Powder -Mono-ammonium phosphate. Added to this, is the
particulate material which is given a special treatment to provide proper
flow capabilities, resistance to caking and provide resistance to packing. It
stops combustion by interrupting the chain reaction sequence and absorbs
the heat; thus smothering the fire by blanketing it. Dry chemicals are less
contaminative and extremely economical. However, for high-tension
electric installation such as a transformer, excellent insulation to prevent
any damage that the DCP may occur. The dry chemical powder is stored
on the drone in limited quantity through a fire extinguishing grenade as
compressed dry powder with compressed nitrogen as propellant; similar to
the one used in DCP fire extinguishers. It is dispelled through a release
mechanism which covers the entire region of the flame.The design is
similar to the current industry models of Fire Ball Extinguishers by Elide.
Each grenade weighs 1.3kgs. These grenades are replaceable after use
and can be easily deployed to completely douse a flame in a 5 meter
radius without harming any crew members in that area. 7 SONIC FIRE
EXTINGUISHER This apparatus is also called as wave extinguisher and
works by employing acoustic waves to suppress the flame. The sound
waves are focused in a specific direction instead of spreading them.In
reality, sound waves have the potential to control oxygen and burning
material. If these two get separated, the fire dies away.The extinguisher
only emits sound, making it ideal to use around equipment and personnel.
The vibration separates the air from the fuel, causing a momentary lapse
in the chain reaction, and then starves the flame from oxygen and hot
vaporized fuel. j12] The waves move the air around them in such a way
that they disrupt the rapid oxidation at the core of a fire. The process

https://api.turnitin.com/newreport_printview.asp?eq=0&eb=0&esm=0&oid=1821191474&sid=0&n=0&m=2&svr=51&r=50.48832845325519&lang=… 4/7
4/26/22, 10:07 PM Turnitin

essentially separates the oxygen from the fuel source and quickly
extinguishes the fire. The AI readily transmits video feed to the control
room for crew to keep track of progress. The self- learning programming
paradigm also causes for auto-alignment and fighting the fire. Internet of
Things (IoT): It is a system of interrelated objects, animals or people,
computing devices, mechanical and digital machines that are provided with
unique identifiers and are capable of transferring data without requiring
the interaction of human with the computer.It consists of the following
components: a. Sensors/Devices: In the first step, the data is collected
from the environment through sensors which can be bundled together to
perform separate functions. b. Connectivity: In this step, the data is sent
to the cloud (storage). This can be done by connecting the sensors to the
cloud through a variety of methods including: cellular, satellite, WiFi,
Bluetooth, low-power wide-area networks (LPWAN), or Ethernet. c. Data
Processing: Once the data is uploaded to the cloud, the software Figure 7:
Route Map followed by Sonic Wave Extinguisher during operation The
frequency used is between 30Hz to 60Hz which therein poses no risk to
humans or machinery nearby. The principle behind the extinguisher is
simple: as they are mechanical pressure waves that cause vibrations in
the medium in which they travel, sound waves have the potential to
manipulate both burning material and the oxygen that surrounds it. If the
sound could be used to separate the two, the fire would be starved of
oxygen and, accordingly, would be snuffed out. performs the required
processes on it. It can be as complex as identifying objects on a given
video. d. User interface: The information thus acquired is made useful to
the user through an interface. The vitals of a given system can be
analyzed by the user through the said interface. Apart from this, the user
can also perform an action to affect the system. Some actions are also
performed by the interface automatically, and the output of the same is
given to the user. Smart Rescue Equipment Accessories Vitality Band:A
band which monitors vitals, including pulse rate, blood pressure which
primarily helps in monitoring the location of the crew. This band can be
strapped on the biceps/forearm under the clothing and is made such that
it does not hinder the working ability of the user. Figure 8: Working of a
GPS system The device includes a motion or fall sensor and will send an
alert to the control room and the nearest drone, if a fall is detected or if
there is an irregular detail of a crew's vitals. The band also has distress
button that can be used to send the signal manually.The drones are first to
arrive at the scene and send live video feed to control room as well as
assess the situation. The Al's continual assessment also recommends
actions to be taken. For example, if someone has fallen and hurt their leg -
the software recommends that the rescuers carry first aid and a stretcher,
if possible.j14] a. Construction: Just like the drone, this transponder can
be made of Carbon Fiber, making it light weight as well as heat resistant. A
strap can be used to attach it to the biceps/arm of the crew. b.Location:
The device in its folds contains a micro GPS (Global Positioning System)
tracking chip such as SKG09BL by SKYLAB. This chip is capable of sending
and receiving GPS signals from the satellites. The output of the chip can
be wirelessly fed to the Local Area Wi-Fi network of the ship and hence to
the AIS. The AIS then alerts the drones and rescue operation is conducted
automatically. The device contains an SOS button which can be tapped in
case of an emergency. This information is immediately transferred to the
AIS which, in turn, activates the drones to reach the said location. As the
drone is already equipped with a projection mapping of the ship and the
input of the position tags, it provides a relatively quicker response and the
AIS decides further course of action.The same method is used when the
person has to be escorted to the nearest exit in case of a smoke-filled
compartment, a blackout or an engulfing fire. c. Measurement of Vitals: An
optical vital sensor by ROHM can be used for measuring pulse rate, blood
pressure as well as stress due to its high-speed sampling of 1024 Hz. It
has a high detection accuracy and low power consumption of 0.44 mA. It

https://api.turnitin.com/newreport_printview.asp?eq=0&eb=0&esm=0&oid=1821191474&sid=0&n=0&m=2&svr=51&r=50.48832845325519&lang=… 5/7
4/26/22, 10:07 PM Turnitin

uses an integrated FIFO (First in First Out) memory to reduce power

consumption. It uses supply voltages if 2.5 to 3.6 volts. It also consists of
an infrared sensor for sensing as to when the device is being worn. The
noise to the circuitry is reduced using optical filters. Fall detection is done
with the help of sensors such as gyroscopes and accelerometers which are
constantly detecting crew's movements. The sensors can detect the
changes in motion as well in height as they are sampled up to 100 times
per second.The output of these devices can be fed along with the tracking
information. Clothing Transponder: The same transponder device can also
be attached onto the clothing. In case of uneasiness or to gain the rest of
the crew's attention, the transponder can be used to send out distress
signal. The drones are first to respond. In both applications of the drone,
the seafarer's lives are rescued through automation and technological
advancements of the drone. In case of rescue from a fire, the drones use
the rescue equipment to locate the seaman trapped in the compartment
and then find the safest possible escape route, whilst fighting the fire and
clearing the pathway for escape. 8 MAINTENANCE 8.1 Charging The
drones use wireless charging, which is a docking station on which the
drone automatically sits and self-charges upon requirement. Wireless
charging works on the principle of induction. The charging dock works as a
primary coil that constantly traverses current. The device to be charged
Figure 9: Wireless Drone Charging Pad contains a secondary coil that is
induced and thus charges it. These pads can be kept at routine places of
use such as every level of the engine room, bridge, etc. 8.2 DCP
Replacement Dry Chemical Powder grenades can be automatically replaced
by the drone through a self-refilling system wherein the routine code
insists that empty grenade pockets are first filled up after every charging
or completion of rescue, whichever is earlier. The QR Code on every
grenade is scanned just before it is placed onto the deployment pocket,
and its replacement date is automatically fed to the system, along with
other details such as information of grenade, manufacturer and all other
requirements of any Life Saving Appliance. Scope The project can be used
on Manned and Unmanned Ships, with usability also inclusive of cruise
liners, coastal vessels and even be extended to ports. On Completely
autonomous Unmanned Vessels, the Drones will work in tandem with the
AI onboard, as extensions. The drones can also be used toaid other jobs
that relate in entering enclosed spaces, working aloft, etc. by keeping the
control room and OOW informed of recent developments in real- time.
Through remote control, these can also help in more frequent and faster
inspections of a greater area onboard. Benefits The proposal requires no
major structural changes and primarily uses the ships pre- existing
systems only small installations are required. Due to greater
maneuverability and speed through flight, response time is significantly
reduced. It also aids in watch keeping and monitoring of systems, as well
as, in quickly spotting errors and issues with machinery and working
through better available video surveillance. 9 CONCLUSION The
assortment of ideas when put together creates a safer, more automated
format of fire-fighting and personnel caregiving which can account for
every person who is at risk and also ensure they are also brought out of
the risk safely. Multiple drones, and each stationed close to a command
centre (Bridge, Engine Control Room, and Accommodation) or places of
high fire risk, can aid in rescuing people and even extinguishing fire before
the first responders show up. Through the Internet of Things, there is
greater availability for surveillance of harm, better assessment of risk, in
addition to the very quick response of efficient rescue systems. Here-in
are discussed widely used systems that have already been developed and
the possible modifications that can be brought in these developed systems
for their application on- board. The life of the seafarers is paramount and
hence, the technological advancement through augmentation in Fire
Fighting techniques with Artificial Intelligence is indispensable to ensure
safety of life followed by safety of property and environment.

https://api.turnitin.com/newreport_printview.asp?eq=0&eb=0&esm=0&oid=1821191474&sid=0&n=0&m=2&svr=51&r=50.48832845325519&lang=… 6/7
4/26/22, 10:07 PM Turnitin

https://api.turnitin.com/newreport_printview.asp?eq=0&eb=0&esm=0&oid=1821191474&sid=0&n=0&m=2&svr=51&r=50.48832845325519&lang=… 7/7