PROJECT PROGRESS REPORT

ON

SMART SURVEILLANCE SATELLITE ‘PRAHARI’

Submitted

In Partial Fulfillment of the Requirements

for the Degree of

Bachelor of Technology
In
ELECTRICAL ENGINEERING

By

Shubhang Shukla (210105061)

Gaurav Pratap Singh (210105075)

Arunendra Pratap Singh (210105020)

Final Year B. Tech Electrical Engineering

DEPARTMENT OF ELECTRICAL ENGINEERING

SCHOOL OF ENGINEERING

H.B.T.U KANPUR

PROJECT GUIDE PROJECT GUIDE

Prof. Dr. C.N. Singh Prof. Dr. Yaduvir Singh

EED, HBTUK EED, HBTUK

1
 CERTIFICATE

This is to certify that the project report entitled “Smart Surveillance Satellite ‘PRAHARI’ “is

submitted by Shubhang Shukla, Gaurav Pratap Singh, Arunendra Pratap Singh in partial

fulfillment of the requirements for the award of the degree of Bachelor of Technology in

Electrical Engineering Department under my guidance and supervision.

Dr. C N Singh
Professor,
Electrical Engineering Department,
H.B.T.U. Kanpur

2
 ACKNOWLEDGEMENT
We want to express gratitude to our project mentors Prof. Dr. Yaduvir Singh, Professor,

Electrical Engineering Department, HBTU Kanpur, and Prof. Dr. C N Singh, Professor,

Electrical Engineering Department, HBTU Kanpur for allowing us to work under their

supervision. Their support, knowledge, and guidance have been very helpful for the progress

of our project.

Shubhang Shukla (210105061)

Gaurav Pratap Singh (210105075)

Arunendra Pratap Singh (210105020)

3
 OBJECTIVE

 Real-time monitoring and decision-making-

Real-time control and decision-making enable instant responses using

machine learning and automated alerts for rapid action.

 Disaster response-

Provide real-time data for rapid response such as forest fire, and mob

gathering.

 Automation and AI integration-

Use AI algorithms for automated, threads or changes in monitored areas.

4
 TABLE OF CONTENTS

Title 1

Certificate 2

Acknowledgement 3

Objective 4

List of Figures 6

List of Table 7

Chapter 1: Introduction 8-10

Chapter 2: Hardware and Software Components

2.1. Quadcopter(X-Frame) 11-13

2.2. Propellers 13-14

2.3 BLDC Motor 15

2.4 Speed Variables or ESCs 15-16

2.5 Flight Controllers 17-19

2.6 Power Distribution Board 19-20

2.7 SHDL Network 23

2.8 Multitask Cascading CNN with MIROF 24

Chapter 3: Mathematical Model 25-31

Chapter 4: Simulation 32-33

Chapter 5: Real World Application 34-36

Conclusion 37

References 38-40

5
 LIST OF FIGURES

Table Page No.

Figure 1.1: ModalAI’s Development Drone for PX4 GPS-

Denied Flight and Obstacle Avoidance- VOXL m500 8

Figure 1.2: The Different Types of UAS’s 10

Figure 2.1: Quadcopter X-Frame 11

Figure 2.2: Types of Quadcopter Frame 12

Figure 2.3: Axis Of Quadcopter 12

Figure 2.4: Sections of a Propeller 13

Figure 2.5: Profile of a Propeller 14

Figure 2.6: Electrical Circuit of a BLDC Motor 15

Figure 2.7: Electronic Speed Controller 16

Figure 2.8: Flight Controller 19

Figure 2.9: ScatterNet Pose Detection Image and Illustration 23

Figure 2.10: Density Map obtained by geometric adaptation of

Gaussian kernel. (a)Input Image (b) Corresponding density map 24

Figure 3.1: Propeller Rotational Motion 26

Figure 3.2: Rotational Control of Quadcopter 30

Figure 4.1: Simulink Model of Quadcopter 32

Figure 4.2: Altitude Control Result of Quadcopter 33

Figure 5.1: Workflow of the Proposed System 36

6
 LIST OF TABLES

Table Page No.

Table 2.1: Component’s Cost 21-22

7
 CHAPTER 1:

INTRODUCTION AND LITERATURE REVIEW

1.1 Literature Review

An IoT-based smart surveillance system for high-security areas leverages interconnected

devices and sensors to enhance security and monitoring. It integrates cameras, motion
detectors, and environmental sensors, all connected to a central system via the Internet of
Things (IoT).[1] The system can detect unusual activities, track movements, and provide real-
time alerts to security personnel. AI integration can also analyze patterns, recognize faces, and
automate responses like lockdowns or alarms. This setup ensures continuous, remote
monitoring, data collection, and faster response times, improving security and reducing human
error.[2]
YOLOv3 (You Only Look Once, version 3) is a real-time object detection model widely used
in smart surveillance systems. It detects multiple objects in an image or video frame by dividing
the image into grids and predicting bounding boxes with class probabilities. YOLOv3 is fast
and efficient, making it ideal for real-time applications like smart surveillance, where it can
detect various objects (people, vehicles, etc.) in live video feeds [3]. Its high-speed
performance and accuracy allow for instant threat detection and monitoring, contributing to
more responsive and intelligent security systems.[4]
A smart surveillance system using machine learning enhances security by analyzing video and
sensor data to detect, recognize, and respond to potential threats. It leverages machine learning
algorithms to identify patterns, detect unusual activities, recognize faces, and differentiate
between objects (like people, vehicles, or animals). Over time, the system improves its
accuracy by learning from new data [5]. It provides real-time alerts, automates monitoring
tasks, and reduces the need for constant human supervision, making security more efficient and
reliable in sensitive or high-risk areas.[6]
A survey of video surveillance systems in smart cities examines the integration of advanced
technologies like IoT, AI, and machine learning to improve urban security and monitoring.
These systems use interconnected cameras, sensors, and analytics to provide real-time
monitoring, detect incidents, and enhance public safety.[7] Smart surveillance in cities supports
traffic management, crime prevention, crowd control, and emergency response by analyzing
video feeds and detecting anomalies. The survey highlights trends such as the use of facial
recognition, predictive analytics, and data integration, showcasing how smart surveillance
systems help build safer, more efficient urban environments.[8]
AI-based satellite surveillance systems for forest fire detection utilize satellite imagery
combined with artificial intelligence to monitor vast forest areas and detect fires early. AI
algorithms analyze real-time satellite data to identify fire indicators like smoke, heat, or
changes in vegetation [9]. These systems can quickly detect potential fire outbreaks, often
before they are visible to the human eye, and provide timely alerts to authorities. AI enhances
the accuracy and speed of detection, enabling faster response and reducing the impact of
wildfires, thus playing a crucial role in environmental protection and disaster management [10].

8
A surveillance satellite with deep learning for traffic monitoring leverages satellite imagery
and advanced deep learning algorithms to track and analyze traffic patterns over large areas.
Deep learning models can automatically detect vehicles, measure traffic density, and identify
congestion, accidents, or unusual patterns [11]. This satellite-based system provides a real-
time, wide-area view of road networks, enabling authorities to monitor traffic flow, manage
traffic more effectively, and optimize urban planning. It offers a cost-effective, scalable
solution for traffic monitoring, especially in remote or hard-to-access regions.[12]
Deep learning-based real-time satellite surveillance for environment monitoring, uses satellite
data and advanced neural networks to track and analyze environmental changes in real-time.
Deep learning models process large volumes of satellite imagery to detect patterns such as
deforestation, pollution, climate change impacts, or natural disasters. These systems can
identify subtle changes in land, water, and atmospheric conditions, providing timely insights
for environmental protection. Real-time processing enables quick responses to emerging issues
like wildfires or floods, helping in disaster management, conservation efforts, and sustainable
resource management [13].
Smart satellite surveillance systems for disaster management, use advanced technologies like
AI, machine learning, and IoT to monitor and respond to natural disasters. These systems
provide real-time data through satellite imagery, detecting early signs of disasters such as
hurricanes, floods, wildfires, or earthquakes. By analyzing environmental changes, they help
predict and track the spread of disasters, enabling faster, more efficient emergency responses.
Smart satellite surveillance supports disaster preparedness, risk assessment, and recovery
efforts by offering wide-area, continuous monitoring, and accurate situational awareness,
helping to minimize damage and save lives [14].
It presents the design and development of an autonomous drone system using IoT technologies.
The drone is intended for real-time surveillance and disaster monitoring, with a focus on
enhancing situational awareness during emergencies [15].

1.2 Introduction

Nowadays, security drones, which are UAVs, are used for various operations and functions

like monitoring, filming, etc. They use advanced and newer technologies like GPS, sensors,

and cameras to inspect and navigate.

Unmanned Aerial Systems (UASs), or drones, continue to increase in capabilities and

sophistication across various applications as they have high mobility, are easily deployed, and

are capable of real-time monitoring of crowd behavior and in-forest monitoring with a focus

on control systems and detection algorithms. They utilize multi-sensor-based detection and

remote sensing of objects. The drone's capabilities are improved via telemetry integration,

9
sophisticated autopilot systems, and an emphasis on accuracy, autonomy, and real-time

communication.

It is based on the concept of leveraging drone technology together with Artificial Intelligence

(AI) and Machine Learning (ML) methods to produce a UAS system that can assist in the

monitoring and assessment of crowd behaviors during peaceful and non-peaceful events, and

in-forest monitoring along with any unusual activity occurring.

Figure 1.1: ModalAI’s Development Drone for PX4 GPS- Denied Flight and Obstacle

Avoidance- VOXL m500

To address this limitation, a potential solution is to employ a 360◦ camera mounted on an

Unmanned Aerial Vehicle (UAV). This combination offers enhanced flexibility and mobility,

allowing users to explore the environment and move around objects within the scene. By

leveraging the mobility provided by the UAV, in addition to 360◦ video, a viewing experience

of up to six degrees of freedom (DoF) can be achieved. This advancement holds promise for

diverse applications like remote video surveillance, scientific exploration, autonomous

manufacturing assistance, agricultural monitoring, and more. However, to fully realize the

potential of these applications, it is crucial to maintain a seamless and responsive interaction

between the user and the UAV by ensuring a more natural viewing experience with accurate

control. This requires ODV to be delivered with high Quality-of-experience (QoE), to provide

10
a truly immersive experience through real-time control of the UAV. Specifically, the high-

quality 4K resolution videos need to be transmitted with ultra-low End-to-end (E2E) latency

(preferably below 1 sec. [2]). However, achieving these metrics over contemporary 5G

networks is highly challenging due to the higher data volume of ODVs, compared to

conventional Two-dimensional (2D) videos. For instance, a High-efficiency Video Coding

(HEVC)-encoded 8K (ultra-high-definition) video typically requires target bitrates ranging

from 20-80 Mbps [3], significantly exceeding the typical throughput of 20 Mbps for UAVs

when operating in the presence of ground users [4], [5]. Furthermore, achieving under one

second of Glass-to-glass (G2G) latency is inherently challenging. This is because a 30 frames-

per-second video encoded with a Group of pictures (GOP) size larger than 16 inherently incurs

a G2G latency of at least one second. However, reducing the size of the GOP negatively

impacts compression efficiency. Further, the intrinsic non-stationarity of the UAV-to-ground

wireless channel and limited computational and energy resources of UAVs amplify these

challenges for UAV-based real-time immersive video streaming.

11
Figure 1.2: The Different Types of UAS’s

12
 CHAPTER 2: HARDWARE AND SOFTWARE COMPONENTS

2.1 Quadcopter (X Frame)

Quadcopters are a type of multirotor drone that has four rotors arranged in a square

configuration. Quadcopters are popular among hobbyists and professionals alike due to their

stability, maneuverability, and ease of use. They are often used for aerial photography,

mapping, surveying, inspection, and other applications that require precision and control.

There are several types of quadcopter chassis or frames, each with its unique design and

characteristics.

Figure 2.1: Quadcopter X-Frame

Here are some common types of quadcopter chassis:

• X-frame: This is the most common type of quadcopter chassis, with arms forming an “X”

shape. It is easy to assemble and maintain and provides good stability which has been used for

the project.

• H-frame: This type of quadcopter chassis has arms that form an “H” shape. It provides better

stability and maneuverability than the X-frame but is also more complex to build and maintain.

• X-Hybrid: This is a combination of X-frame and H-frame designs. It offers a balance between

stability and maneuverability and is popular for both beginner and advanced pilots.

Spider frame: This type of quadcopter chassis has arms that extend outward from a central

hub, resembling a spider’s legs. It is lightweight and flexible, making it ideal for

acrobatic maneuvers.

13
 Figure 2.2: Types of Quadcopter Frame

2.2 Aerodynamic principles of functioning of Quad-copter

Figure 2.3: Axis Of Quadcopter

Pitch: The movement on the lateral axis that bites or ascents is denominated as pitch.

Roll: The rotation around the longitudinal axis essential to abandon a rectilinear flight

the path is called roll.

14
Yaw: The movement around the vertical axis is called yaw motion.

The flight principles governing quadcopters are rooted in aerodynamics and physics.

They utilize four rotors, each spinning in opposite directions, to create lift and

maneuver through the air.

2.3 Propellers

The propeller is the most important element among all the electronics of a multirotor team

because each propeller is a rotating wing.

The propeller is responsible for converting the power supplied by the engine into useful

work for the flight.

A propeller is formed by a set of aerodynamic profiles like those presented by the wing

of an airplane and with a variable angle from the center of the propeller to the tips. In

each section of the propeller, the thrust effect is produced.

Figure 2.4: Sections of a Propeller

The following figure shows the aerodynamic profile of the cross-section of a propeller

at 75% of its diameter. The highest edge is called the leading edge and the lowest

edge is the trailing edge.

15
 Figure 2.5: Profile of a Propeller

2.4 PROPELLER SPECIFICATIONS

A propeller is defined by two numerical parameters: its diameter in inches and its pitch.

For example, 9*5, and 10*4.7 are actual propeller configurations.

The passage of a propeller can be defined as what would advance an aircraft in

horizontal flight when it completes one revolution. That it is a 9*5 propeller indicates with

its 5” pitch that the airplane flying propelled by it advances 5” forward in each

revolution.

Propeller 1045 is used in this project which means it has 10 inches diameter and 4.5

inches pitch.

16
2.5 BLDC MOTOR

The Brushless DC Motor presents several advantages. -higher torque/weight ratio -less

operational noise -longer lifetime -lesser generation of electromagnetic interference. The only

limitation is its inherent heat generation, which it transfers to the external environment by

conduction.

Figure 2.6: Electrical Circuit of a BLDC Motor

Where,
U = voltage applied
Ra = armature’s resistance
Vb = back electromotive force
Parts of a BLDC motor =
1. Rotor - one which rotates
2. Stator - one which is stationary

2.6 Speed Variables or ESCs (Electronic Speed Controller)

An electronic speed controller (ESC) is a device that controls the speed and direction of an

electric motor, typically a brushless DC (BLDC) motor. An ESC takes a signal from a radio

control receiver or another control source and converts it into a pulse-width modulated (PWM)

signal to control the motor’s speed and direction.

An ESC typically consists of a microcontroller, power electronics, and a user interface. These

are the following parts and functions of ESCs:

1. Output filter capacity- It is used to eliminate voltage oscillations.

2. Set of 3 vertical branches of legs named FET - These 6 switches are low-

resistance channels MOSFET-JEFT transistors

17
3. FET driver circuitry-this component is simply an element that adapts the output of

the microcontroller to the voltage and current levels.

4. Microcontroller- it is the program governing the transistors

5. BEC= Battery Elimination Circuit, which is an electronic circuit that reduces the voltage

of the batteries.

The Electronic Speed Controllers is the interface between the control and power stage

of brushless electric motor. It receives signal from a low-power microprocessor and

converts it into a 3-phase high-power alternating signal

While selecting an ESC, the following variables must be taken into account-

1)Maximum current it can handle- 30A continuous, 40 A for 10 seconds

2)No of cells for which it is designed- 2-4 cells Li Po

3)Inclusion or not of BEC - 5V, 3A for external receives and servos

4)Weight- 32gms

Here we are using 30A ESC for our project.

Figure 2.7: Electronic Speed Controller

18
2.7 FLIGHT CONTROLLERS

The flight controller is the brain of a drone and is responsible for stabilizing the aircraft,

maintaining its orientation, and controlling its movements in the air. The flight controller uses

sensors to measure the drone's orientation, speed, and altitude and then adjusts the drone's

motors to maintain stability and perform desired maneuvers. Here are some of the key functions

of a flight controller in a drone: Stabilization: The flight controller uses accelerometers and

gyroscopes to measure the drone's orientation and adjust the motor speeds to keep it level and

stable in the air. Altitude control: The flight controller uses a barometer to measure the drone's

altitude and adjusts the motor speeds to maintain a constant altitude. Position hold: Some flight

controllers have GPS receivers that allow the drone to maintain a fixed position in the air, even

in windy conditions or when the pilot is not actively controlling it.

Autonomous flight: Some flight controllers have advanced features that enable the drone to fly

autonomously, such as waypoint navigation, follow mode, and automatic return-to-home. The

flight controller receives commands from the pilot or a ground control station and translates

them into motor commands to control the drone's movements. The flight controller also

provides telemetry data, such as battery voltage, GPS location, and altitude, to the pilot or

ground station for monitoring and control.

19
2.8 KK 2.1.5.5 FLIGHT CONTROLLER

The KK2.1.5 flight controller is widely used in the drone industry, particularly for applications

such as drone spraying. It offers a range of features and functionalities essential for stable and

precise flight control. Designed by Rolf Bakke, the KK2.1.5 is known for its simplicity and

reliability. It supports various multi-rotor configurations including quadcopters, hexacopters,

and octocopters, making it versatile for different drone spraying set-ups. One of its key features

is its large LCD screen, which provides real-time data and settings adjustments without

additional equipment or software. This feature simplifies set-up and tuning, even for users with

minimal programming knowledge. The KK2.1.5 offers easy calibration for various

components including ESCs (Electronic Speed Controllers), ensuring smooth and accurate

throttle response. Additionally, it includes built-in self-leveling and stabilization features,

which are crucial for maintaining operations, especially in windy conditions. With its

integrated gyro and accelerometer sensors, the KK2.1.5 provides precise attitude and altitude

hold, essential for maintaining consistent spraying patterns and avoiding collisions.

Furthermore, the KK2.1.5 supports firmware upgrades, allowing users to benefit from

improvements and new features over time. Its open-source nature also fosters a supportive

community of users who share tips, troubleshooting advice, and customizations. In summary,

20
the KK2.1.5 flight controller is a robust and user-friendly choice for drone applications,

offering reliability, versatility, and ease of use for both beginner and experienced operators.

Figure 2.8: Flight Controller

2.9 POWER DISTRIBUTION BOARD

The power distribution board (PDB) used in the KK2.1.5 flight controller for drone spraying

is a vital component responsible for managing power distribution to various parts of the drone

system. It serves as a central hub where power from the battery is routed to the flight controller,

motors, and other onboard electronics. The PDB typically features multiple solder pads or

connectors for connecting the battery, ESCs (Electronic Speed Controllers), and other

peripherals. It ensures efficient power delivery, reducing the risk of voltage spikes and

providing stable power to all components. In the KK2.1.5 flight controller set-up, the PDB

21
interfaces directly with the flight controller, allowing it to monitor battery voltage and current

consumption. This information is crucial for maintaining safe operation and preventing over-

discharge.

Moreover, the PDB may include additional features such as voltage regulators or BECs

(Battery Eliminator Circuits) to provide regulated power to auxiliary devices like cameras,

lights, or telemetry systems. Its compact design and robust construction make it suitable for

drone applications, where space and weight are critical considerations. Additionally, the PDB’s

layout and configuration are optimized for easy installation and maintenance, ensuring reliable

performance.

22
 TABLE 2.1 COMPONENT’S COST

Item Part Name Purchase Link Dimensions Weight(g) Price

No.
1. Bonka 11.1V 3300mAh 35C 3S https://robu.in/product/bo 137*43*20 260 2569
Lithium Polymer Battery nka-3300mah-35c-3s-
lithium-polymer-battery-
pack/
2. FlySky FS-i6 2.4G 6CH https://robu.in/product/fl 47*26.2*15 14.9 4459
PPM RC Transmitter With FS-iA6B y sky-fs-i6-2-4g-6ch-
Receiver ppm-rc- transmitter-with-
fs-ia6b- receiver/

https://www.flyrobo.in/fl
ys ky_fs-
i6_2.4g_6ch_afhds_rc_tr
an smitter?search=fs-
i6&description=true

3. DJI 2212 920KV Brushless DC https://robu.in/product/or 28*28*46 60 600*3=180

Motor for Drone with Black Cap ange-hd-propellers- 0
(CW Motor Rotation) 104510x4-5-abs-dji-
black- 1cw1ccw-1pair-
premium-
quality/
4. 2212 920KV Brushless DC Motor https://robu.in/product/22 28*28*46 60 600*2=120
for DJI Silver Cap (CCW Motor 12-920kv-brushless- 0
Rotation) motor- dji-red/

5. Propellers 1045(10X4.5) ABS DJI https://robu.in/product/or 44 94*4=

Black 1CW+1CCW-1pair ange-hd-propellers- 376
104510x4-5-abs-dji-
black- 1cw1ccw-1pair-
premium-
quality/
6. 30A BLDC ESC https://robu.in/product/sta 45*24*9 23 388*4=
ndard-30a-bldc-esc- 1552
electronic-speed-
controller/

https://www.flyrobo.in/30
a
_brushless_motor_esc_fo
r
_airplane_quadcopter?sea
r
ch=esc&description=true

23
7. KK2.1 Flight Control Board https://robu.in/product/kk 51*51*12 21 5979
2-1-5-multi-rotor-lcd-
flight- control-board-
with- 6050mpu-and-
atmel- 644pa/

https://www.flyrobo.in/kk
2
.1.5_lcd_flight_controll_b
o
ard_for_fpv?search=kk2.
1& description=true
8 100A Multirotor ESC Power https://robu.in/product/m 50*50*2 120
. Distribution Battery Board ultirotor-esc-power-
distribution-battery-board-
quadcopter/
https://www.flyrobo.in/esc
-multirotor-power-
distribution-battery-board-
for-quadcopter-multi-axis-
model?search=power+distr
ibution+board&description
=true

Jumper wire,heat shrink,

ziptye,DST, Soldering kit, nylon
band

24
 SOFTWARE TO BE USED

2.7 ScatterNet Hybrid Deep Learning (SHDL) Network

ScatterNet Hybrid Deep Learning (SHDL) Network, which is ScatterNet, can then take the

human pose estimations and analyze the positions and movements. The system detects violent

individuals in real time by processing the drone images in the cloud. The figure (left) illustrates

the 14-body key points annotated on the human body. The illustration on the right shows the

skeleton figure corresponding to the humans in an image. The angles (right image, shown in

green for a few limbs) between the various limbs in this structure are used by the support vector

machine (SVM) to recognize the humans engaged in violent activities. After the poses are

determined, the orientations between the limbs of the estimated pose are used by the SVM to

identify individuals involved in violent activities. [4] The intent is to incorporate ScatterNet

Hybrid Deep Learning Network methods into this AI-UAS platform.

Figure 2.9: ScatterNet Pose Detection Image and Illustration

25
2.8 Multitask Cascading CNN with Multiscale Infrared

Optical Flow (MIROF)

Visible-light-based object detection and understanding methods are often greatly influenced by

the environment. As a result, a single type of feature derived from aerial monitoring videos is

often insufficient to characterize variations among different abnormal crowd behaviors. To

address this, we propose combining two types of features to better represent behavior, namely,

multitask cascading CNN (MC-CNN) and multiscale infrared optical flow (MIR-OF),

capturing both crowd density and average speed and the appearances of the crowd behaviors,

respectively. The Infrared (IR) camera was paired with the NVIDIA Jetson TX1 for the infrared

vision system. Since there are no published infrared-based aerial abnormal-behavior datasets

the researchers are providing a new infrared aerial dataset named the IR-Flying Dataset, which

includes sample pictures and videos of different scenes of public crowd behavior.

Figure 2.10: Density Map obtained by geometric adaptation of Gaussian kernel. (a) Input
Image (b) Corresponding density map

26
 CHAPTER 3: MATHEMATICAL MODEL

The quadrotor, UAV (Unmanned aerial vehicle) is an autonomous unmanned flying machine

offering us an additional degree of freedom and an unlimited workspace, many researchers are

facing this new device, and several studies have led to the discovery of the areas of application

of quadrotors Modelling and control of quadrotor has been studied in several papers;

mathematical and dynamic model in [8, 9] linear and nonlinear controller in [2]and for a

quadrotor control with PID controller [10, 1]but the most of them is not complete modeling in

all cases of quadrotor situation; the recent research is mainly based on the fact that the quadrotor

is an element of agent set to schedule a mission that requires cooperation and the control of a

group of the quadrotor, several research is based on the complete modeling with more

mathematical details [4, 7] This paper presents complete modeling that is required for projects

that use the quadrotor as an elementary agent for cooperation. this paper is organized as follows

first, a mathematical model of quadrotor is presented.

Figure 3.1: Propeller Rotational Motion

27
3.1 DYNAMIC MODEL

The quadrotor UAV model is defined in the two frames, an inertial reference frame E: {e1, e2,

e3} and a body fixed frame B: {b1, b2, b}.

Two vectors associated with the quadrotor, q for position and rotation in the E frame and V for

translational and rotational velocity in the body-fixed frame where X mass of the UAV and is

the relative position of the center of Roll Pitch Yow Her orientation of the quadrotor in &

Frume, with the constraint that

The Roll φ motion is given by the thrust imbalance between f2andf4 forces around, we can

write the rotation matrix R(φ) as:

1 0 0
𝑅𝜙 = [0 𝐶𝑜𝑠𝜙 −𝑠𝑖𝑛𝜙]
0 𝑠𝑖𝑛𝜙 𝐶𝑜𝑠𝜙

The Pitch & motion is given by the thrust unbalance between f{1} and f{4} around by. the

Rein given by:

𝐶𝑜𝑠𝜃 0 𝑠𝑖𝑛𝜃
𝑅𝜃 = [ 0 1 0 ]
−𝑠𝑖𝑛𝜃 0 𝐶𝑜𝑠𝜃

The Yaw motion thrust imbalance of all so moments around the axis, the rotation matrix is:

𝐶𝑜𝑠𝜓 −𝑠𝑖𝑛𝜓 0
𝑅𝜓 = [ 𝑠𝑖𝑛𝜓 𝐶𝑜𝑠𝜓 0]
0 0 1

We can now define the global rotation matrix R of the quadrotor in the inertial frame & with

the product of three matrices as R = R{2}*R{b}*R{d}

𝐶𝑜𝑠𝜃𝐶𝑜𝑠𝜓 𝑠𝑖𝑛𝜃𝐶𝑜𝑠𝜓 − 𝐶𝑜𝑠𝜙𝑠𝑖𝑛𝜓 𝐶𝑜𝑠𝜙𝑠𝑖𝑛𝜃 𝐶𝑜𝑠𝜓 + 𝑠𝑖𝑛𝜙𝑠𝑖𝑛𝜓

𝑅 = [ 𝐶𝑜𝑠𝜃𝑠𝑖𝑛𝜓 𝑠𝑖𝑛𝜙𝑠𝑖𝑛𝜃𝑠𝑖𝑛𝜓 + 𝑐𝑜𝑠𝜙𝑠𝑖𝑛𝜓 𝑐𝑜𝑠𝜙𝑠𝑖𝑛𝜓 − 𝑠𝑖𝑛𝜙𝑐𝑜𝑠𝜓]
−𝑠𝑖𝑛𝜃 𝑠𝑖𝑛𝜙𝐶𝑜𝑠𝜃 𝐶𝑜𝑠𝜙𝐶𝑜𝑠𝜃

The relation between and is given by the following equation

𝑅 = ̇ 𝑅Ω
̂

28
Where, R(φ), R(θ), R(ψ), and R matrices of rotation are defined in the special orthogonal group

SO(3), which have the following properties:𝑅 3 → 𝑆𝑂(3)

𝑥̂𝑦 = 𝑥 × 𝑦

Where,

0 −𝑥3 𝑥2
𝑥
𝑥̂ = [ 3 0 −𝑥1 ]
−𝑥2 𝑥1 0

To express the angular velocity of the quadrotor in the fed is

∨
𝛺 = (𝑅 𝑇 𝑅̇ )

The equation becomes

∨
0 1 ∨
0 ∧
0 𝜙
𝛺 = (𝑅𝜑𝑇 𝑅𝜃𝑇 [ ] 𝑅𝜃 𝑅𝜙 + 𝑅𝜙𝑇 [𝜃̇] 𝑅𝜙 + [ ] )
0 0
𝜓 0
0

We use this property

𝑝 1 0 −𝑠𝜃 𝜙̇
𝛺 = [𝑞 ] = [𝜌 𝐶𝜙 𝑠𝜙𝐶𝜃 ] [ 𝜃̇ ]
𝑟 0 −𝑠𝑝 𝐶𝜙𝐶𝜃 𝛹̇

DOW write the relation between iq and V in the matrix form as:

𝑞̇ = 𝑇. 𝑉

The dynamics equation of the unmanned aerial vehicle UAV

m𝑣 = 𝑓𝑅𝑒3 ̇ − 𝑚𝑔𝑒3

M=JΩ + Ω̇ × JΩ

The translational and rotational kinetic energy of the quadrotor is given by

1
𝑇𝑡𝑟𝑎𝑛𝑠 = 𝑚̇𝑋 𝑇 𝑋̇
2
1 𝑇
𝑇𝑟𝑜𝑡 = 𝛺̇ 𝐽𝛺̇
2

The inertia of the quadrotor in the body-Esed frame we assume is symmetric.

29
The potential energy of the body is

U=mgz

Given hy F = FX + FY and ta ∗ u

= [[tau0 , tau0 , taupsi ]] 𝑇 𝑤ℎ𝑒𝑟𝑒 𝐹𝑋 𝑖𝑠 𝑡ℎ𝑒 𝑏𝑦 𝑝𝑟𝑜𝑝𝑒𝑙𝑙𝑒𝑟𝑠 𝑎𝑛𝑑 𝐹𝑡 𝑖𝑠 𝑡ℎ𝑒 𝑑𝑟𝑎𝑔 𝑓𝑜𝑟𝑐𝑒𝑠 𝑎𝑝𝑝𝑙𝑖𝑒𝑑 𝑡𝑜 𝑡ℎ𝑒

𝑞𝑢𝑎𝑑𝑟𝑜𝑡𝑜𝑟 𝑎𝑙𝑜𝑛𝑔 𝑒𝑎𝑐ℎ 𝑣𝑒𝑐1 , 𝑣𝑒𝑐2, 𝑣𝑒𝑐3 𝑣𝑒𝑐𝑡𝑜𝑟𝑠 , 𝑓 − 𝑠𝑢𝑚 𝑖 = 1 𝑡𝑜 𝑘 𝑓𝑖 𝑎𝑛𝑑𝐹𝑋 − 𝑓𝑅 ∗ 𝑒𝑆

̇̇
𝐼𝑥 𝜃𝜓̇(𝐼𝑥 − 𝐼𝑦 ) 𝐽𝑟 𝜃Ω̇ 𝑟 𝑟𝜙
[𝐼𝑦 ] = 𝜙𝜓̇ ̇ (𝐼 − 𝐼 ) + [−𝐽 𝜙Ω𝑟̇ ] + [ 𝑟𝜃 ]
𝑥 𝑧 𝑟
𝐼𝑧 𝑟𝜓
̇ 0
[ 𝜃𝜙̇(𝐼𝑦− 𝑖𝑥 ) ]

The quadrotor is defined as six degrees of freedom, three for the position of the center of mass,

and three for rotation, which represents the output variables and also the final state of the

quadrotor in three-dimensional space, but the system’s input is just four, one thrust force and

torque, we named by Ui =1,...,4 variable inputs of UAV controller. (15)

4

𝑈1 = ∑ 𝑓1 = 𝑓1 + 𝑓2 + 𝑓3 + 𝑓4 = 𝑏(𝜔12 + 𝜔22 + 𝜔32 + 𝜔42 )

𝑖=1

3.2 ROTOR DYNAMICS

𝑑𝑖
𝑣 = 𝑅𝑖 + 𝐿 +𝑒
𝑑𝑡

We have the electric and mechanic power of the motor are equal, therefore Km=Ke, and were

placed (Jp ηr2 +Jm) by Jt we obtain from:

𝐾2 𝑑 2 𝐾
𝜔̇ 𝑚 = − 𝑅𝐽𝑚 𝜔𝑚 − 𝜂𝑟 3 𝐽 𝜔𝑚 + 𝑅𝐽𝑚 v
𝑡 𝑡 𝑡

30
 Figure 3.2: Rotational Control Of Quadcopter

For this equation control, we used the Taylor series in the first order at a value to linearize:

2𝑑𝜔0 ̇ 𝐾𝑚2 𝐾𝑚 𝑑𝜔02

𝜔𝑚 = − ( + ) 𝜔 + 𝑣 +
𝜂𝑟 3 𝐽𝑡 𝑅𝐽𝑡 𝑚 𝑅𝐽𝑡 𝜂𝑟 3 𝐽𝑡

3.3 QUADCOPTER CONTROL

3.3.1 Position control

From the dynamic equations, we have a decoupled system where the translational accelerations

don’t depend on angular acceleration. From the first three equations, we can extract the pitch

and roll angles when we introduce the desired position and attitude.

From the equations above we can calculate and depend of position vector and attitude

𝐾𝑓𝑥 ̇ 𝐾𝑓𝑦
(𝑥̈ + 𝑚 𝑥)𝑐𝜓 + (𝑦̈ + 𝑚 𝑦)𝑠𝜓̇
tan 𝜃 =
̈ 𝐾𝑓𝑧
𝑧 + 𝑔 + 𝑚 𝑧̇

31
 𝐾𝑓𝑥 ̈ 𝐾𝑓𝑦 ̇
(𝑥 + 𝑚 𝑥)̇𝑠𝜓 + (𝑦̈ + 𝑚 𝑦)𝑐𝜓
sin 𝜙 =
2 2 2
𝐾 𝐾̈ 𝐾
√(𝑥̈ + 𝑓𝑥 𝑥̇ ) + (𝑦 + 𝑓𝑥 𝑦̇ ) + (𝑧̈ + 𝑔 + 𝑓𝑥 𝑧̇ )
𝑚 𝑚 𝑚

In our case, we assumed that the disturbance is neglected, then K fx =K fx =K fz =0 we study

that in other

𝑥̈ 𝑐𝜓 + ̇ 𝑦̈ 𝑠𝜓̇
tan 𝜃 =
𝑧 +̈ 𝑔

(𝑥𝑠𝜓 +̈ 𝑦̈ 𝑐𝜓̇
sin 𝜙 =
√(𝑥̈ )2 + (𝑦̈ )2 + (𝑧̈ + 𝑔)2

32
CHAPTER 4: SIMULATION

Figure 4.1: Simulink Model of Quadcopter

33
Figure 4.2: Altitude Control Result of Quadcopter

34
 CHAPTER 5: REAL WORLD APPLICATION

Smart surveillance satellites are highly versatile tools for enhancing security and monitoring

activities at local events or within educational institutions. Below are some real-world

applications for these specific environments:

1. Public Events (Concerts, Festivals, Sports, Parades)

 Crowd Monitoring and Control: Smart satellites can provide an aerial view of large
crowds, helping organizers and law enforcement monitor the event in real time.
Satellites equipped with AI can analyze crowd behavior, detect unusual patterns (like
sudden movements or stampedes), and alert authorities to take preventive measures.
 Perimeter Surveillance: For large events, smart satellites can patrol the perimeter to
detect unauthorized entries, breaches in security, or potential intrusions. They can also
monitor parking lots and surrounding areas for suspicious activity, improving overall
security.
 Traffic Management: Satellites can help manage traffic flow around the venue,
providing live data on congestion or accidents, allowing authorities to redirect traffic
or deploy resources more efficiently. They can monitor pedestrian zones as well,
ensuring smooth and safe foot traffic.
 Emergency Response: In case of emergencies, such as medical incidents, satellites can
provide a quick assessment of the situation by reaching crowded or remote areas where
human responders may struggle to navigate. They can help pinpoint the exact location
of the issue, reducing response time for medical or security personnel.
 VIP Protection: Drones equipped with facial recognition and AI technology can track
and monitor VIPs or high-profile individuals at events, enhancing security measures by
providing a live feed of their surroundings.

2. Educational Institutes (Schools, Colleges, Universities)

 Campus Surveillance: Drones can monitor large campuses, offering a real-time aerial
view of school grounds. They can patrol areas that are less frequented or are difficult
to monitor with stationary security cameras, such as remote sections of the campus,
parking lots, or recreational areas.
 Emergency Situations: In the case of an emergency, such as a fire, natural disaster, or
even an active shooter situation, drones can provide real-time surveillance to guide first
responders, locate individuals in need of help, and offer situational awareness. This can
be especially valuable for large educational campuses.
 Monitoring Access Points: Drones can be used to monitor the entrance and exit points
of educational institutions, ensuring that only authorized personnel and students are on

35
 campus. AI-equipped drones can recognize vehicles or individuals and detect
unauthorized access.
 Campus Events Monitoring: During large campus events, such as sports games,
graduations, or open days, drones can monitor crowds, assist in traffic control, and
provide aerial footage for security purposes. They help ensure safety by keeping an eye
on large gatherings and identifying potential risks in real time.
 Vandalism and Theft Prevention: Smart drones can be deployed to monitor campus
buildings and other facilities after hours to prevent theft, vandalism, or unauthorized
access. AI-powered drones can detect suspicious activities, alert security teams, and
provide video evidence.
 Environmental Monitoring: For schools that are focused on sustainability or
environmental education, drones can monitor campus resources like water usage,
energy efficiency, or green spaces. Drones equipped with sensors can collect data on
environmental conditions, such as air quality, that can be used for research and
educational purposes.

3. Sports and Recreational Activities

 Field and Facility Monitoring: For sports events held on school campuses or public
fields, drones can monitor the facilities for safety and security. They can assist in
organizing large gatherings or sports meets by providing a real-time overview of the
area, ensuring that spectators and participants are in designated zones.
 Live Broadcasting and Event Footage: Drones can also be used for live streaming or
recording aerial footage of school events, such as football games, tournaments, or
performances. This footage can be shared with parents, alumni, or the community and
can also be used for security monitoring.

4. Disaster Drills and Safety Training

 Simulation Monitoring: In educational institutions, smart drones can be used during

emergency drills (fire, earthquake, lockdown drills) to monitor how students and staff
respond. They provide real-time footage and can help in evaluating the effectiveness of
evacuation procedures and safety measures.
 Safety Audit and Infrastructure Monitoring: Drones can inspect buildings for
structural issues or monitor construction work on campus, ensuring safety standards are
met. They can quickly assess roofs, walls, or outdoor facilities that may be hard to reach
by human inspectors.
Smart surveillance drones bring numerous benefits for both public events and educational

institutions by enhancing security, improving emergency response, and providing valuable data

for better decision-making.

36
Figure 5.1: Workflow of the Proposed System

37
 CONCLUSION
Till now we have completed the basic drone infrastructure and altitude simulation of the

drone. Looking forward we have to complete the programming of the flight controller and set

up the transmitter and receiver to make the flight of the drone possible.

38
 REFERENCE
1. O. I. D. Bashi, W. Z. W. Hasan, N. Azis, S. Sha e, and H. Wagatsuma Autonomous quadcopter

altitude for measuring risky gases in the hazard area, J. Telecommun., Electron. Comput.

Eng., vol. 10, nos. 25, pp. 3134, Jul. 2018.

2. H. T. Berie and I. Burud, Application of unmanned aerial vehicles in earth resources

monitoring: Focus on evaluating potentials for forest monitoring inEthiopia,

Eur.J.RemoteSens.,vol.51,no.1,pp. 326335, Jan. 2018.

3. S. Barbarino, O. Bilgen, R. M. Ajaj, M. I. Friswell, and D. J. Inman, A review of morphing

aircraft, J. Intell. Mater. Syst. Struct., vol. 22, no. 9, pp. 823877, Jun. 2011.

4. M.J.G.Guarnizo, R.C.L.Trujillo, and M. J. A. Guacaneme, Modeling and control of a two DOF

helicopter using a robust control design based on DK iteration, in Proc. 36th Annu. Conf. IEEE

Ind. Electron. Soc. (IECON), Nov. 2010, pp. 162167.

5. Z.-C. Qin, Y. Xin, and J.-Q. Sun, Dual-loop robust attitude control for an aerodynamic system

with unknown dynamic model: Algorithm and experimental validation, IEEE Access, vol. 8,

pp. 3658236594, 2020.

6. G. Kavuran, A. Ates, B. B. Alagoz, and C. Yeroglu, An experimental study on model reference

adaptive control of TRMS by error-modi ed fractional order MIT rule, J. Control Eng. Appl.

Informat., vol. 19, no. 4, pp. 101111, Dec. 2017

7. D. Erdos, A. Erdos, and S. E. Watkins, "An Experimental UAV System for Search and Rescue

Challenge," IEEE Aerospace and Electronic Systems Magazine, vol. 31, no. 3, pp. 40-46, 2016.

8. A. K. Sahu, A. Prakash, P. Anurag, and S. Pandey, "Drone-Based Smart Surveillance System

for Disaster Management," International Journal of Innovative Technology and Exploring

Engineering (IJITEE), vol. 8, no. 9S3, pp. 743-748, 2019.

39
9. M. N. Islam, S. Abdullah, and A. Dey, "Smart Drone Surveillance for Real-Time Object

Detection," Proceedings of the International Conference on Computing, Communication,

and Intelligent Systems (ICCCIS), IEEE, 2020.

10. H. S. Goel and M. K. Rana, "Autonomous Drone for Smart Surveillance and Disaster

Response," International Journal of Engineering Research & Technology (IJERT), vol. 9, no. 5,

pp. 1023-1027, 2020.

11. M. A. Hasan, M. M. Alam, and M. N. Anwar, "Aerial Surveillance Using Smart Drones: A

Review on Technologies and Applications," International Journal of Advanced Computer

Science and Applications (IJACSA), vol. 12, no. 2, pp. 278-285, 2021.

12. S. V. Jose and K. P. Mani, "Design and Development of an Autonomous Drone for

Surveillance," International Journal of Recent Technology and Engineering (IJRTE), vol. 8, no.

1S4, pp. 1162-1166, 2019.

13. F. Jiang and H. Adeli, "Drone-Based Smart Surveillance in Unstructured Environments Using

Deep Learning," IEEE Transactions on Neural Networks and Learning Systems, vol. 31, no. 8,

pp. 2984-2995, 2020.

14. T. Q. Duong, H. D. Tuan, and H. V. Nguyen, "Aerial Surveillance Using UAVs for Emergency

Management and Security," IEEE Access, vol. 7, pp. 121011-121024, 2019.

15. D. B. Alves, M. A. Corrêa, and P. H. R. Mota, "Autonomous UAV Surveillance System Using

Deep Learning-Based Object Detection," Proceedings of the 2019 International Conference

on Unmanned Aircraft Systems (ICUAS), IEEE, 2019.

16. N. Kumar, A. Choubey, and V. K. Agarwal, "AI-Powered Drone for Real-Time Smart

Surveillance," International Journal of Scientific and Technology Research (IJSTR), vol. 9, no.

2, pp. 3965-3970, 2020.

17. L. Jayakumar and R. Prakash, "Autonomous Surveillance Drone with Computer Vision and

AI," International Journal of Advanced Research in Computer and Communication

Engineering (IJARCCE), vol. 8, no. 10, pp. 180-185, 2019.

40
18. H. K. Vo, T. D. Ngo, and T. T. Pham, "Smart Surveillance Using UAVs with Real-Time Face

Recognition and Tracking," International Journal of Electrical and Computer Engineering

(IJECE), vol. 10, no. 4, pp. 4462-4469, 2020.

19. A. Anitha, A. M. Vino, and S. Suganya, "Drone Surveillance System Using Image Processing

for Object Detection and Tracking," Journal of Theoretical and Applied Information

Technology (JATIT), vol. 97, no. 4, pp. 1167-1178, 2019.

20. V. V. Varadarajan and V. Y. Latchoumi, "Real-Time UAV Surveillance System with AI-Powered

Analytics," International Journal of Innovative Research in Computer Science & Technology

(IJIRCST), vol. 7, no. 4, pp. 234-240, 2019.

21. S. Agrawal, K. M. Sharma, and A. K. Sharma, "Design of an IoT-Based Autonomous Drone for

Surveillance and Disaster Monitoring," International Journal of Advanced Science and

Technology (IJAST), vol. 29, no. 5, pp. 5641-5650, 2020.

41
APPENDICES

42