CARGO DELIVERY SYSTEM WITH DRONE

GAZİ UNIVERSITY ENGINEERING FACULTY

ELECTRICAL-ELECTRONICS ENGINEERING DEPARTMENT

EEE492 ELECTRICAL-ELECTRONICS ENGINEERING DESIGN 2

RESULT REPORT
2020-21 Spring Semester

GROUP :
1- Emrullah ÜLKER (151110072)
2- Oğuzhan BULUT (151112008)
3- Utku ÇELİKER (151112009)
4- Oğuz Kaan ULUIŞIK (161110071)

COUNSELOR:
Dr. Öğr. Üyesi Mehmet KARAKAYA

18 JUNE 2021 Ankara

iii
CONTENTS
Page

SYMBOL AND ABBREVIATIONS …………………………………………………...……….vi

1. PROBLEM STATEMENT….……………………………………….………………..……........1
1.1. Need Statement…………………………………………………………………………1
1.2. Objective………………………………………………………………………………..1
1.3. Background And Related Works……………………………………………………….2

2. REQUIREMENTS SPECIFICATION…………………………………………………………..4
2.1 The Requirements………………………………………………………………………4
2.1.1. Mission Requirements………………………………………………………..4
2.1.2. Design Requirements…………………………………………………………4
2.2.Constraints………………………………………………………………………5
2.2.1 Sustainability………………………………………………………………….5
2.2.2 Health, Environment and Safety………………………………………………5
2.2.3 Legal…………………………………………………………………………..5
2.2.4 Ethic…………………………………………………………………………...5
2.2.5 Design Constraints…………………………………………………………….6
2.3 Standards………………………………………………………………………..6
2.4. Evaluated Concepts and Configurations………………………………………..6
2.5. Calculation and Amount of Unit Weights……………………………………...7
2.6. Final Conceptual Design Configuration………………………………………..7

3.DESIGN…………………………………………………………………………………………..8
3.1 Level 0…………………………………………………………………………………...8
3.2 Level 1…………………………………………………………………………………...8
3.2.1 Dimensional Parameters of Final Design…………………………………...14
3.2.2 Structural Features and Capabilities of the Final Design…………………...14
3.3. Level 2 and Device Selection …………………………………………………………14

iv
4. DESIGN VERIFICATION……………………………………………………………………..24
4.1 Test Results…………………………………………………………………………….24
4.1.1 Testing the GPS……………………………………………………………….24
4.1.2 Testing the ESC and Motors………………………………………………….25
4.1.3 Testing the FCB………………………………………………………………29
4.1.4 Testing the DSP…………………….…………………………………………30
4.2. Verifications…………………………………………………………………………...32
4.2.1 Design Constraints……………………………………………………………32
4.2.2 Mission Requirements………………………………………………………...33
4.2.3 Design Requirements…………………………………………………………33
4.3. Standarts………………………………………………………………………..34

5. SUMMARY AND CONCLUSIONS..........................................................................................35

6. ADMINISTRATION ………………………………………..…………….…..………...……..36
6.1. Team Organization……………………………………………………………………36
6.2 Timeline………………………………………………………………………………..37
6.3 Work packages………………………………………………………………………...37

7. TEST PLAN…………………………………………………………………….……..………..38

REFERENCES.………………………………………………………..………..…….………...…41

ADDITIONS………………………………...……………………………..…………..………….42

v
 SYMBOLS AND ABBREVIATIONS
The symbols and abbreviations used in this study are presented below with their
explanations.
Symbols Descriptions

0
C Santigrat Degree
cm Santimeter
gr Gram
m Meter
m² Meter square

Abbreviations Descriptions

A Amper
DSP Digital Signal Processor(Dijital sinyal işleyici)
ESC Electronic Speed Controller(Elektrik Hız Kontol)
FCB Flight Control Board(Uçuş kontrol kartı)
GPS Global Positioning System(Küresel konum sistemi)
KV Volt başına rpm
Li-Po Lityum-Polimer Batarya
mAh Mili amper hour(Miliamper -saat)
Prop Propeller(pervane)
Rpm Repeat per minute(Dakika başına devir)
TL Turkish Lira
UAV Unmanned Aerial Vehicle
V Volt

vi
1. PROBLEM STATEMENT
This detailed design report gives detailed information about the design process followed in
the design stages of the project on "Cargo Delivery System with Drone" of the "EEE492"
course.
The main problem addressed in the project is that shipping systems are too dependent on
human beings, so the accidents and mistakes encountered can negatively affect human,
product and environmental factors.
Job descriptions and rules determined by our consultants in the design of the transport
system, today's scientific studies in this field, the necessity of having an appropriate and
affordable cost, and the environmental factors in the environments where the use of the
vehicle (drone) is aimed have played a major role.

1.1. Need Statement

In our daily life and business life, the quality of the work to be done in order to achieve
quality is as important as the fast and safe transportation of the products to be produced.
In today's classical cargo transportation; The main problems such as damage to the product
during transportation, minor or major damages to living and non-living environmental
factors during transportation, unacceptably prolongation of the transportation time, the
efficiency of the product transportation process to remain at certain levels and therefore the
inadequate energy saving. For this reason, health and safety issues are evaluated as a
design problem and appropriate solutions are tried to be presented.

1.2. Objective
The main purpose of the design is to minimize the human factor in product transportation
and to contribute to the industry by increasing the occupational safety in transportation
systems.The design of the vehicle has been made in accordance with the user goals, taking
human health and environmental factors, dangers into consideration. The project also aims
to ensure that the vehicle reaches its target in the shortest possible way in a healthy way,
thus providing fast and easy use and increasing energy savings.

1.3 Background And Related Works

Since this project addresses the design of an autonomous drone system that can fulfill the
requirements, previous studies have been investigated and several projects have been found
the project to be designed. Two of the academic studies on similar fields are given in the
following titles together with the summary information from the thesis studies.

Gps Guided Autonomous Drone

In this project; An autonomous flying drone, especially a quadcopter, is being designed.
The drone is equipped with a GPS tracking system and is programmed to fly autonomously
from one location to another using GPS coordinates. Safety and durability have been taken
into consideration considerably due to the possibility of collisions with various objects. In
addition to collisions, the drone is also rugged enough to operate in moderately windy
conditions. The aim of the project is to establish a small-scale autonomous air delivery
system similar to the future delivery system that Amazon is targeting. The top and side
views of the final version of the project are given in Image 1.

Image 1. Side and Top

view of the Final Project

Automatic Route Determination And

Tracking For Drones

In this project, it is aimed that the drone
(UAV) follows the assigned route and
returns to its starting point without any
need for an operator. For this purpose, it is aimed to be operated automatically through the
GPS module and sensors to be added to the device.
Basic physical components for the operation of the drone; defined as energy source,
controller, sensors and modules. In this study, PixHawk flight controller, which is widely
used as a controller, and a 5200mAh 3S Li-Po battery was chosen as the energy source.
GPS module to determine the position of the drone, ultrasonic sensors to detect obstacles,
barometer to determine the height, and flow sensor to ensure the stability of the drone in
the air. In the application, DroneKit library was used to develop the necessary software to
control the drone, and Mission Planner was used to work with the flight controller to
monitor the instant sensor and status information of the device. Two different operating
environments, simulation and real systems, were designed to test whether autonomous
flights operate successfully and to perform the flight. Open source ArduCopter virtual
drone was used in the simulation environment. The simulation was installed in Windows

10 environment and a virtual simulation drone was created by compiling ArduCopter using
MavProxy in the Mavlink infrastructure. The created virtual simulation drone is monitored
through Mission Planner. Image 2 shows the route the drone follows.

Image 2. Drone’s Route

2.1. Requirements
2.1.1. Mission Requirements
The tasks that the drone will undertake are defined as follows in order to meet the need and
to provide problem solving;
Task 1 - Cargo handling: To ensure the delivery of the product; It must be able to carry the
cargo at the starting point and take it to the desired delivery area.
2. Task - Automatic routing: Delivery should be able to reach the desired point
automatically.
3. Mission - To follow the route without dropping the package: It is expected to deliver the
package without being affected by weather conditions such as wind and rain during
delivery.
4. Mission - To be able to land at the delivery point: When reaching the delivery area, it
should be able to land near the delivery point.
Mission 5 - Returning to the starting point: After the delivery is completed, the vehicle is
expected to return to the starting point.
Task requirements are summarized in Table 1.

2.1.2. Design Requirements

The design of the vehicle should be made to meet the mission requirements. For this;
• Selecting the engine to be selected with a power suitable for the load to be carried and the
weight of the vehicle
• Selecting the batteries that are strong enough to keep the vehicle in the air until the
vehicle leaves the starting point and returns to the starting point,
• Choosing the appropriate GPS module for the drone to determine the route.
• Selecting the engine and the appropriate ESC to control the speed adjustments of the
vehicle engines,
• Selection of propellers suitable for the engine and with a diameter that can lift the cargo,
• Choosing the appropriate motherboard for the integration of sensors and control units
• There should be a suitable cargo platform in order to prevent the cargo from being
affected by wind and rain.

2.2. Constraints
2.2.1 Sustainability
Sustainability is the ability to continue the project even after it is completed, in other
words, ensuring the permanence of the project. Updating the electronic devices and
software used in the project ensures the sustainability of the project.
Health and environmental criteria have always been prioritized in the project. Personal
protective equipment was used in the use of solder, heater, and cutting tools during the
assembly phase. The health and environmental damage risks that may be caused by the
product during operation were examined and reduced to zero. Occupational Health and
Safety criteria were used as a provider tool in these three issues.

2.2.3 Legal
Persons who damage living, non-living things and the environment as a result of using the
design outside of its intended purpose and not being used under the specified standards and
conditions will be legally subject to material and moral penal sanctions. In addition, the
manufacturer will face legal sanctions as a result of the failure to perform the tests and
trials required during the production of the product and this situation endangers the
primary person and the environment. In addition, as a result of unauthorized copying and
production of the electronic and mechanical design of the vehicle, our patent owner team
has the right to initiate a material and moral legal struggle against the real or legal persons
who perform this transaction. This legal process may result in the other party's
compensation or imprisonment.

2.2.4 Ethic
During the project process, attention was paid to act in accordance with originality and
ethical principles. The topics such as showing the sources used in the research, awareness
of responsibility, confidentiality and security are handled in accordance with ethical rules.
In addition, group members always acted with courtesy and respect towards each other.

2.2.5 Design Constraints

1) The developed vehicle should reach the coordination with the GPS module with the
lowest error rate and energy expenditure on the shortest route.
2) The engines of the vehicle must be suitable for lifting 5 * 5 * 5cm loads.
3) The package should be left within an area of 1 m².
4) After the drone has delivered the cargo, it should return to its starting point
autonomously.
5) The system cost should be limited to a maximum of 3000 TL.
6) Electronic circuits must be able to function fully between -20 ° C and + 40 ° C
IEEE802.11p /LTE-V2V : It covers all vehicle communication systems as an
umbrella protocol where sensor data can be transferred at high speed with low
loss.

ISO 45001: It is the standard of occupational health and safety.

ISO 27001: It is designed to protect information assets and to provide adequate

and proportional security controls that give confidence to interested parties.

IEC 61850: It is an international industrial communication protocol used for

electrical network data communication.

2.4. Evaluated Concepts and Configurations

Our design is designed to be simple and similar to other designs in the same space.
Efficiency is prioritized and task-oriented work is aimed. All the specified criteria have
been evaluated and the appropriate hardware and software have been selected to complete
the task completely. Its maneuverability and speed capability has been adjusted according
to the equipment and the route layout. It is aimed to return to the starting point in case of
any malfunction or obstacle. Vehicle equipment has been selected in accordance with its
ease of operation under the specified conditions.Unlike other existing projects, it was
predicted that the change we made in the load lifting mechanism would bring along a few
problems and it was decided to place appropriate equipment on the drone by producing
appropriate solutions to these problems. With the adjustments, tests and experiments to be
made, the vehicle is expected to complete the task completely in accordance with all
criteria.

2.5. Calculation and Amount of Unit Weights

The list of materials to be used in the design is given in Table 2 below. The weights of
these materials are approximately indicated next to them.

Metarial name Amount (Adet) Mass (gr)

Frame(Drone Çerçevesi) 1 160
Brushless DC Motor 4 4x40
propeller 4 4x7
ESC 1 30
DSP 1 50
FCB 1 200
Li-Po Baterry 1 115
Camera 1 20
Load Platform 2 50
 Approximately total system mass (all existing) 823 gr  1000gr

2.6. Final Conceptual Design Configuration

Simplicity, understandability and durability have been taken as a basis in the design in
accordance with the determined project constraints and the tasks to be performed by the
vehicle. In order to facilitate the maneuvers to be made by the vehicle, the necessity of the
motors to be used to operate with maximum efficiency by consuming the least energy and
the necessity of staying in the air sufficiently were realized by using brushless motors.
In the selection and design of the electronic equipment, priority was given to reducing the
cost and weight load on the vehicle, so it was thought that it would be appropriate to use
this loss in the transport of the load by minimizing the loss of movement caused by the
weight of the vehicle. In other words, with the reduction of the weight of the vehicle, it is
aimed to carry more weight by using less power. This directly affects the cost of the power
supply to be used and the efficiency of the vehicle by increasing the load carrying.
Due to the advantages of its cost and easy programming capability, compatibility with
sensors (camera, GPS) and motor drivers to be used, the number of available digital pins,
and its low power consumption, it can be used at low voltage values. It was designed by
taking.

3.DESIGN

3.1 Level 0

Digital Signal
camera Proccessor Battery
(DSP)

Communication
Control unit
unit Power unit

Land control Motors and

sensors
propellers
The basic modules that should be approximately available in the autonomous drone are
shown in the figure. In addition, the basic connection scheme of these elements is
discussed. Connections or elements may change at the next design level. It is even possible
to experience the number of additional elements used.
In addition, in this connection diagram, the input output output, current and power data of
electronic or electromechanical systems are not specified. Only their relationship with each
other is given at an introductory level.
3.2 Level 1
At this design level, we will look at what is in the blocks we saw in the previous stage and
which individual elements they should be connected to, and what data they will receive
from them or what data they will transmit to them.

Camera
The system needs image data as it will operate on a computer vision basis. Therefore, the
camera is an essential component of this system. At this stage, it was decided to work with
a 480p camera in order not to take much memory. The camera would only send data to the
DSP module. Data is digital data.

480p
Vision data DSP
Camera

DSP
DSP is the unit that will process the data it receives from the camera. At this stage, it was
decided to use a raspberry pi electronic card for the DSP unit.
This card, which will be programmed in Python language, will process the image data and
transmit it to the central control unit in the most appropriate format.

DSP Processed data Control unit

The battery will serve as the power source for the drone. When deciding on its
size and type, the tare calculation of the drone should
be chosen according to the current to be drawn
depending on the rotor RPM value required
depending on the weight of the package it will carry.
In addition, since other circuit elements work with
electrical energy, these elements can be energized
separately or can be provided over the main battery.
Reduces energy

Energy Other
Power unit
Battery components

Power Unit
The power unit is responsible for transmitting the
necessary power to the motors for the movement to take
place. In drones, this element is called ESC (Electronic

Speed Controller). ESC or ESCs must be compatible with

the body of the drone. For a drone with a short wingspan,

the central ESC called 4-in-1 should be used, or 1-1 or 2-1
ESCs can be used for a drone with a long wingspan. The
biggest advantage of 4-in-1 ESC is that it is a single element and is located in a safe place
on the body.
ESCs have essential 2 input and 1 output terminals. One of the input terminals is connected
to the power supply and one to the control unit. Thus, the ESC adjusts the power to be
transmitted to the motor according to the signal from the control unit. Output terminal is
connected only to the motor.
Reduces
Battery Energy ESC
energy Propeller

Energy
control
signal

Control unit

Propeller And Motors

Drone engines should be selected with the capacity to produce the necessary torque to
3100 KV motor. This means that this motor is rotating at 3100 rpm without any propellers
attached under 10 V / 6A voltage and current. It has been decided to use a 2 pole 6 inch
diameter propeller as the propeller.

ESC power

Communication Unit
The communication unit is responsible for the
communication between the ground control
center and the drone, as well as the location
determination by using coordinates since it
contains a GPS module. Here, the ground
control center is intended as a control. It detects simple commands that will come from the
control and transmits them to the control unit. At the same time, the battery fill rate, height,
engine rotation speed, etc. From the sensor group. It
is also responsible for sending the data to the
controller.
It was decided to use NRF24 module for bi-
directional communication within the unit. The
communication band is 2.4 GHz and the
communication between two stations can go up to 250 meters in an unobstructed
environment. The module can work integrated with Ardunio. In addition, PCB design can
also be made.

Sensors Control Unit

Remote Communica
Remote Control tion Unit

The controller represents a ground control station.

Commands such as “cancel task”, “return” are
transmitted to the drone by the transmitter from the remote.
The remote also informs the drone operator about the
physical condition of the drone by reflecting the sensor data
sent by the drone’s sensors with the help of the communication unit with the display
indicator. In this way, the user can be sure that everything is working well on the drone.
Sensors
The drone has to use some data while performing its mission. Some of these are data such
as height, battery charge rate, engine rotation speed. Our goal is to ask the drone to send us
the data that will answer questions such as how
accurately we are following the route we
envisioned while doing the drone mission, is it at
the height it should be right now, will its battery
be enough to complete the mission.
The sensors that should be on the drone at this
stage are:
1) Altimeter (for altitude information)
2) Gyroscope (for directional information)
3) Optical encoder sensor (to measure motor speed)
4) Ultrasonic proximity sensor (to avoid collision)
5) Battery charge sensor (for battery information)

Control Unit
The most important part of the drone is its brain, the control unit. This unit is called the
“Flight Control Board (FCB)”. Market review has shown that some types have built-in
gyroscope and GPS modules.
Flight control card is the unit that receives data from sensors, evaluates them and
makes final decisions. This unit also evaluates the commands that will come from the
control and implements them immediately.
Since our drone will be processing the image, it takes the processed image data
from the DSP module and whether it is where it should be, or whether it sees the signs or
images that will enable it to perform the task, whether to approach depending on this,
decide whether to reduce the height or do any of these, cancel the task and return to the
starting point. Takes the necessary decision for.
Since the flight control card works in conjunction with the ESC, the flight control
card also monitors the speed of the motors and drives the ESC to change the speed data
where necessary.
Communication ESC
unit

FCB

Land control sensors

3.2.1 Dimensional Parameters of Final Design

The design was designed by considering the weight and the required lifting force, as it
contains many sensors and equipment. The length of the vehicle planned to cover the
design requirements was designed as approximately 300 mm. Considering the weight of
the equipment owned by the vehicle, it was thought that these equipment should be placed
in suitable places for the vehicle. Since it will lift the load from the front, it is aimed to
place the heavy equipment starting from the rear of the vehicle towards the front in order to
stay in balance. The weight of the vehicle together with the equipment was calculated to be
around 1 kg. In order to pull the load, the pulling force of the platform was considered and
the weight effect of the load on the vehicle was calculated. A design plan was made
considering the sensitivity of the elements that may be affected by vibration while in
motion and any disconnection criteria.

3.2.2 Structural Features and Capabilities of the Final Design

It is requested from the design that loads of certain weight and characteristics are
autonomously transported, left to certain delivery areas and returned to the starting area.
The design was discussed under separate headings as load and vehicle. A GPS module and
image processing sensor have been added to the vehicle so that the vehicle can move
autonomously. USB connection was made to the vehicle over a windows based software in
order to receive commands from the user. It is equipped with sensors in order to easily
locate the vehicle. A return to the starting point was provided upon the need for the vehicle
to receive commands again. A brushless motor was used to make the vehicle move easily.
It was determined that there are some uncertainties in this project, which was designed by
• Uninterrupted communication with GPS, connection status,
• The condition that the vehicle can move easily in weather conditions.

3.3. Level 2 and Device Selection

In this design phase, the finest details of all components will be included, and the exact
parts to be used will be selected while the drone is finalized. On the other hand, the drone
will be assembled and the test phase of the vehicle will begin.
The selection of the subsystems in the design has been designed in accordance with the
criteria of cost, efficiency, easy coding and being part of a user-friendly system.

Raspberry Pi
The first thing to do in the design is to choose a controller that mainly distributes the tasks
to be performed and controls the performance of the tasks. The fact that this controller is
compatible with other subsystems to be used, can operate at low power, is at an acceptable
level in terms of cost, and is easy to program, will add extra advantages to the design.
Arduino family Raspberry Pi
It is included in the microcontroller It is included in the microprocessor family.
family. It contains units such as It is a central processing unit, it can only
processor, memory, input / output, ROM, manage and decide transactions, but cannot
RAM and does not need to be added perform transactions alone. Peripherals must
peripherals. Therefore, it can control and be added from outside.
perform transactions alone.
It is used to control simple systems. It is used for controlling complex systems.
It is open source, easy to program, easy Programming is complex and may take a
to learn and its programming interface is long time to learn. Requires installation of
more user-friendly. It does not require its own operating system.
installing an extra operating system.
It is low cost and easier to obtain. It can It is costly and more difficult to obtain. It is
be used in simple systems to further used in projects where high performance is
reduce costs. required.
Tablo 3. Arduino and Raspberry Pi Comparison
As seen in the table, it was seen that the Arduino family has more advantages over the
Raspberry Pi family. But the tasks to be done by the project are at certain levels, and
Arduino was removed from the options because it was not at a level to meet these
requirements.

DSP
DSP, raspberry pi was chosen in our drone vehicle. This card is of one type and we have
decided to use RASPBERERY-4.
Broadcom BCM2711, quad-core Cortex-A72 (ARM v8) 64-bit SoC @ 1.5GHz
2GB / 4GB LPDDR4-3200 SDRAM
4 / 5.0 GHz 802.11ac supported wireless network, Bluetooth 5.0, BLE supported
2xUSB 3.0, 2xUSB 2.0 ports
2 × micro-HDMI ports (4K 60fps supported)
2-line MIPI DSI display port
2-line MIPI CSI camera port
265 (4kp60 decode), H264 (1080p60 decode, 1080p30 encode)
Graphics processor with OpenGL ES 3.0 support
Micro SD card slot for operating system and data storage
5V power input (minimum 3A) via USB-C
5V power input (minimum 3A) via GPIO connection
50 grams
Important Note: GPIO pins on Raspberry Pi work at 3.3V level. It is not compatible with
5V.

Camera
The system needs image data as it will operate on a computer vision basis. Therefore, we
took care that it is both cheap and high quality. As a result of the comparisons made from
different sales sites on the Internet, it was decided to use the following camera.
ANGL
VIDEO
SYNCHRONIZATIO SIZE(mm RESOLUTIO SENSOR E OF PRIC
SHOOTIN
N ) N TYPE VISIO E
G
N
OmniVisio
5MP 1080p
Compatible with n 54x41 129,9
25x20x9 (2592x1944 720p
Raspberry pi 3-4 OV5647 degree 9 TL
piksel) 640x480p
(5 MP)
Estimated weight is 20 grams

Flight Control Card

The flight control card is the decision-making centerpiece of the drone and is an
indispensable element for the autonomous drone. During this period, we scanned many
flight control cards and found that there are many models in the 150 TL - 1500 TL band.
However, for our autonomous drone project, we decided to use the "PixHawk" card, which
has autoplot software, which will allow us to add or add to this software and has a lot of
 SELLER
WEIG FEATUR INTERNAL
MODEL SIZE PACKAGE INCLUDED AND
HT ES SENSORS
PRICE
32 Bit Package Ichoroscope:
ICM-20689
Arm®
+ PIX
Cortex®- Accelerometer /
+
Pixhawk M7, gyroscope: BMI055 Robotista
+
PX4 44x84x12 15.8 216MHz, n.com
Magnetometer (compass): Security key
v2.4.8 2MB IST8310 599,96 TL
+
memory, +
Barometer (altitude Buzzer
512KB
sensor): MS5611
RAM Prunes
gyroscope: ICM-20689 MINI PIX

+ +
Buzzer
Accelerometer: ICM- +
20689 Neo-M8N Gps Module
+
SMT32f40 +
Black Gps Holder
Pixhawk Barometer: MS5607 +
7 (32bit Banggood
PX4 + Power module
36x36 200 Cortx-M7 + .com
v2.4.8 Compass: IST8310 Minim OSD w / shell
168Mhz Support for up to three + 689,83 TL
Mini I2C Expand Splitter Module
Arm) + +
Compass with built-in PPM Module w / shell
+
+
433 MHZ 100 mw / 500 MW telemetry
2x outer support 2 +
Cable
(maximum)
L3GD20 3 axis digital 16 bit PIX
32 bit 2M gyroscope +
+ Security key
flash
LSM303D 3-axis 14 bit +
memory accelerometer / Flight control shell
Pixhawk magnetometer +
Banggood
STM32F4
PX4 - 37 + Buzzer .com
27 Cortex MPU6000 6-axis +
v2.4.8 accelerometer / From 6 pins to 6 pins
542,92 TL
M4
magnetometer +
168MHZ + 4Pin to 4Pin line
MS5611 high precision +
RAM
barometer 3pin DuPont line

Electronic Speed Controller

Since our drone is not for industrial purposes, but for training purposes and the amount of
load it will carry is not much, it is predicted to have a short wingspan. Since the wingspan
is short, instead of placing 1 ESC in each arm, that is, instead of 4 1-in1 ESCs, it has been
 (V)
1x15A 4in1
ESC
JHEMCU EM15A 2-4S LIPO 15 20 29x31 4.9 - 147,46
1x 6P 1.0
mm cable
1x 12A
GEPRC F411 2-4S LIPO 12 - 23x21 2.7 ammeter 155,46
4-in-1 ESC
1x 15A 4-in-
1 ESC
BLHELI_ Programmable
HAKRC 2-4S LIPO 15 20(5S) 27x31 4.8 1xCable 161,49
S for racing
1x bag of
screws
1x ESC
1x8Px1mm
cable
1000uF/35V
Software
3-6S Filter
RACESTAR AIR50 55 - 30.5x30.5 12.1 support and 172,30
(12-36V) Capacitor
amperage
1x14AWG
Red Cable
1x14AWG
Black Cable
1xESC
JHEMCU EM25A 2-4S LIPO 25 30 29x31 4.9 1x6P 1mm - 172,38
Cable
Support
BLHELI_
LDARC 2-3S 12 15 25x28 4.0 1x ESC voltage 174,65
S
module
BLHELI_
HAKRC 2-4S 20 25(5S) 27x31 3.7 1x ESC programmable 188,05
S
Tryo89-
Programmable
EACHINE Racing- 2-4S 20 23(3S) 30x30 5.5 1x ESC 189,62
for racing
Drone
Includes
current sensor
RACESTAR BLheli_S 3-6S 35 40 36x36 13 1x ESC and use of no 195,81
extra
capacitors

Communication Unit
The communication unit is built into the flight control card. There is no need for an
external communication unit or GPS transceiver.

Land Control
There is no need to design or purchase and compile an external controller. On the contrary,
the interface of the flight control card is suitable for communication with the computer.

Sensors
The sensors that are internally on our flight control card are sufficient for us, and it has
been decided to use the following sensors as an option.
1) Optical reflection sensor (to measure motor speed)
Body

AXLE
SUGGESTI SPECIFICATI
IMAGE WEIGHT(gr) DISTANCE PRICE
ONS ONS
(mm)
4 x Motors
Emax XA2212
1400KV
Brushless Motor
PCB frame board
+
Optimized frame
4 x Hobbywing
design, making
SKYWALKER
20A Brushless the wiring of
ESC ESCs and battery
160 330 + safer and easier 96,53
4 x 8045 +
Propellers Plenty of
+
mounting space
1 x Battery 3S-
for autopilot
4S LiPo
systems.
+
1 x CC3D
Flight
Controller
FC mounting
hole: 30x30mm
118 250 - + 132,50
Camera mounting
distance: 27mm
Flight Control:
APM 2.6 /
Pixhawk 2.4.6
Flight Controller
Engine: EMAX
Can connect the
MT2213 935KV ESC and power
460 450 Engine
225,50
cable directly to
ESC: 20-30A ESC
 Motor

1.motor
48,50TL

Recommended probe: 80xx-90xx

2. motor
52,97 TL

Recommended probe: 6040

3. motor
65,74 TL
77,57 TL

5. motor
90,81 TL

Weight factor is important in motor selection. The calculated weight, including the body of
the drone, was 534 grams. It is predicted to be 600 grams with internal sensors. Although it
is large, the maximum weight of the drone was calculated as 700 grams in case of a 15%
margin of error, and the engine selection was made in this regard and the 1st engine was
won. Depending on the choice of the propeller to be made later and the selection of the
battery, the engine may change, but since the selected engine is well above the drone
weight values, it will not be affected by the weight of the propeller, battery and other
connection elements.
Here; L: length, P: pitch, B: number of palms.
For example, we have a 6 * 4.5 (also known as 6045) propeller. Here 6 represents the
length and 4.5 represents the pitch. Length units are in inches.
To give another example, the numbers 5 * 4 * 3 or 5040 * 3; Represents a 5 inch long, 4
inch pitch, 3 blade or 3 blade propeller.
Sometimes you can see the word "BN" on the impeller. The "BN" here indicates a stubby
nose propeller (bullnose). Sometimes you may see the words "R" or "C" on the impeller.
The expressions here refer to the rotation of the propeller. If you see the word "R" on the
propeller, it means "reversed". It should be attached to the propeller rotating clockwise. If
there is a "C" it must be attached to the motor rotating counterclockwise.As the drone
team, we decided to use the 2-blade 7045 propeller, that is, the propeller with the above
features. The 7 gram weight of the drone is included in the calculated weight, the
calculation is correct. The price of our product is 28,13 TL

Battery
Considering the engine to be used in the drone, the most suitable battery for us
ZOP Power 11.1V 1500mAh 40C 3S Lipo Battery.
Its price is 140.96 TL and its weight is 115 grams.Please pay attention to the operating
voltage of the motor and the operating voltage of the battery.

4.DESIGN VERIFICATION

4.1 Test Results

There are test results in this section.

4.1.1 Testing the GPS

The GPS module is integrated into the PX4 control card. In order to test the PX4 control
card, the GPS test was carried out with the simulation programs we determined. The GPS
module was tested using Gazebosim and utilities Ardupilot and MavLink in a Linux
environment.

By using the location marking and target location identification codes included in the
Ardupilot open source library, the targeted points in the Gazebo Simulation Program were
determined. Afterwards, the coordinates were determined and the program written in
python language shown in Appendix was entered and the program was run. GPS motion
tests of MavLink software with the help of one-to-one scale maps were verified
simultaneously with GazeboSim.
4.1.2 Testing the ESC and Motors

In the shematic above, a BLDC-ECS complex was tried to modeled. Although we didn’t
use one by one components, circuit response it similar to real one. In the simulation
PLECS simulator is used and we thank you Gazi University fort his feature.
First of all we tried to model ESC, which has a same topology as DC-AC inverter. We
used this topology, because BLDC motors are driven sinusodial pulse width modulation
generators and control by changing output frequency. In SPWM, carrier signal changing
the switching of mosfet frequency and referance sinüs signal changes the output frequency.
While referance frequency is increasing, stator magnetic field is changing faster and motor
is rotating faster. However, when output frequency is changed, motor reactance is
changing and thus, phase current are changing.
Graph which is shown at the top side is called sinusodial carrier modulateder signal and the
signal which is state at the bottom side is switching frequency of the inverter mosfets. By
using this method, we can creat more efficient sinusoidal wave forms. These technique
reduces motor’s vibrations.
In BLDC motor controlling, frequency is changed and angular speed is directed
proportional with frequency. However, the most diffucult way of controlling BLDC motors
is : While Frequency and velocity is increasing, current is decreasing. On the other hand,
while drone is operating at high loads, ESC will try produce more frequent switcghing
signal and current so we need a closed loop controller. In SPWM, directly PID using is not
operating correctly but there is more efficient control algorithms.

At motor testing step, we measured the parameters and detected worst conditions. In high
current systems, the most important topic is heating problems. When drone is operating, it
will carry a load and this causes a force, and it means electric current. For prevent burning,
we designed a thermal fuse in PLECS simulator.

When switch is turning off, in PLECS simulator does not turn on the switch at thermal
mode. But er can see that when the current is increasing, the components are heating and
fuse is cutting the circuit
Let say that this ESC is produsig current which causes 105°C. When we decrease the
frequency, motor reactance will decreases and phase current will increase. Thus our
component will be heated by current. If will current flows in a long time, our component
might melt so we must cut the circuit by a fuse in a emergency situation.
4.1.3 Testing the FCB
We used the Pixhawk4 card as the FCB, that is, the flight control board. We aimed to test
GazeboSim and Ardupilot softwares supported by the control card and using python
language.
The Pixhawk4 control card is responsible for the stable performance of the drone's landing-
take-off, speed adjustment and movement tasks. It processes the data from the sensors it
contains and sends this data to its own software, thus adjusting the drone engine speeds.
These tasks have been successfully tested in the simulation software used. The python
codes used are given in Appendix FCB Code.

Preparation of the simulation environment

 Simulation environment login screenshot
4.1.4 Testing the DSP (Image Processing)
To explain our code in detail, we determined HSV value of. We used paint to get HSV
value of the blue which we use in this application. We prefer to use 960x480 pixels for the
frame size.
We choose Gaussian blur to reduce the noise because mean filter (box blur) significantly
affects the values of all pixels also we didn’t choose median filter because of long

calculation time. In Gaussian blur *exp(-x2 /2 σ) ,σ =standard deviation.

Using this formula obtained core matrix which convolve with pixels.
We had HSV values of colors so, we change color format BGR which is default in python
to HSV. Then we made a mask for the color in order to fix errors in the masking process
we used erode operation and dilate operation respectively which is called opening also we
chose number of iteration 2.
We found just external contours because we need only external ones. Then, we created
minimum rectangle which is cover contour to determine the center we used moment
algorithm. The moments of a contour computed using the Green's formula.
Using Deque library we wrote follow up algorithm which follows past value of center.
Finally, we showed center of rectangle. Also there is a line which follows the center’s past
values.
To test our image processing code we drew blue cross shape on paper which is 8x8 cm. We
used small paper so that we could push the limits of our code.
 For 1 meter we used 8x8 cm paper.

For 4 meters we
used 15x10 cm blue
plastic layer because our
camera couldn’t
cath 8x8 cm paper.
For 10 meters we used 15x10 cm blue plastic layer. Although our program works well in
10 meters with 15x10 cm plastic layer, we decided to use 1 m2 blue layer to guarantee the
stable operation of the program.

4.2 Verifications

4.2.1 Design Constraints

1. The GPS module has been chosen in such a way that it can go to the destination with the
least error. In addition, when the energy consumption levels of the GPS module are
examined, it has been observed that it provides the expected energy efficiency. Since GPS
uses satellite communication, it uses the shortest path that has a bird's eye view.
2. The motors were selected taking into account the weights of the drone and the payload.
With the tests carried out, it has been predicted that the motors can lift the specified load
and leave it to the desired point.
3. It has been verified on the basis of image processing tests that the package can be left in
the targeted area using the image processing method with the desired accuracy.
4. Thanks to the software made, after the package is delivered, the drone will use the GPS
module to return autonomously to the point where the delivery started, taking into account
the GPS deviation.
5. The cost calculation was made taking into account the economic conditions of that day.
However, due to the changing economic conditions for today, the cost of the drone exceeds
the predicted limit.
6. All electronic parts have been selected to operate at these temperature values. It has also
been taken into account that motors may overheat during operation.

4.2.2 Mission Requirements

3. The description here has been carried out in conjunction with items 1, 2, 3 and 4 in the
previous section.
4. Thanks to image processing, the drone located near the target will successfully deliver. It
has been understood from the tests that the image processing and GPS work as desired.
5. The job description here is provided with the 4th item in the previous section.

4.2.3 Design Requirements

1. Based on calculations and simulations, it has been understood that the selected engines
can carry the load and the drone itself.
2. Considering the average consumption of the elements, it is understood that the amount
of energy that the battery can store is enough to make the delivery of the drone and return
to the starting point.
3. This item was realized by the GPS module managed by the controller.
4. The electronic speed controller has been selected in accordance with the motors and it
has been learned that they will work in harmony with the factory information.
5. The propellers have been selected according to the recommended dimensions for use in
the engine as well as the overall dimensions and weights of the drone.

As a result, almost all of the desired conditions and constraints can be predicted
through the calculations, assumptions and simulations made. The cost constraint that
cannot be realized is related to economic conditions and varies according to the prices of
electronic elements in the market..

4.3 Standarts

IEEE802.11p /LTE-V2V :
Sensörlerin mevcut olduğu PX4 elektronik kontrol kartı IEEE802.11p /LTE-V2V
standartına uygun üretilmiştir.

ISO 45001:
Drone fiziksel olarak yapım aşamasına geçmediği için güvenlik tedbir ve önlemleri de
alınamamıştır
.
ISO 27001:
Bilgi ve verilerimizin korunması yüklenen antivirüs programları ile sağlanmıştır.
 Haberleşme aracı olarak drone kurulum görev belirleme esnasında bluetooth ve wi-fi
bağlantıları kullanılacaktır. Görev isnasında ise bağlantı mesafesi aşılacağı için herhangi
bir haberşelme cihazı olmayacaktır. Kurulum ve görev belirleme esnasında kullanılan
haberleşme araçları IEEE802.11p /LTE-V2V standartına uygundur.

5.SUMMARY AND CONCLUSION

In this study, it is aimed to develop an autonomous drone for use in cargo delivery. With
the autonomy of the device, the workforce used has been reduced to almost zero. In
addition, accidents that may occur during today's cargo transportation process are
prevented. The vehicle goes to the GPS coordinates entered into it, thanks to the control
card and Python programming, and determines the point where it should leave the load via
the camera and positions itself. Then it leaves the mechanical arm he has to drop the cargo
with the programmed software again. After completing the cargo drop process, it
determines the return route to the starting point (can be the cargo distribution center) and
returns. Units to be used were determined with the tests performed and design plans were
made. Then the units were tested one by one and made ready for integration.
In this testing process, motor and ESC are tested together for safety of the autonumous
drone. It is decided to replace a thermal fuse for drone safety, because of any over load
condition. Through test results, we determined the safe operating interval for the drone
approximately.
At the second part, we performed drone simulation test and we observed that drone can
At the last step, we developed image and video signal processing and the results were
incredibly high rated from the wanted. Responses of the drone is really fast for a real time
operating with video processing part.
After successful completion of the tests with integrated systems, the project was made
ready for completion. During the project construction process, ethical values were
complied with and the design was shaped by considering health, environment and safety.
Accordingly, it is envisaged to be completed in the targeted time in the plans determined
for the project.

6- ADMINISTRATION
The project group consists of four people. It is aimed for each group member to have full
knowledge of the project, and detailed research topics were given to the group members in
line with their interests and abilities. These people and their research topics are given
below.

6.1 Team Organization

Emrullah Ülker: Electronic components and connections are concerned with the
mechanical, structural design to fulfill the vehicle's requirements. With regard to the
personality of the researcher, he provided the literature research related to the project and
the examination of the examples.
Oğuzhan Bulut: In line with his interest in control systems, he is interested in researching
the necessary materials for determining the movement and route of the vehicle and their
integration into the vehicle.
Oğuz Kaan Uluışık: In line with his interest in image processing and software, he conducts
research on the sensor and electronic card that the vehicle will operate. It conducts research
on the platforms where the software will be built..
Utku Çeliker: Due to its familiarity with the market, it conducts market research in order to
achieve the minimum cost of equipment to be used. It focuses on battery, motor, motor
drivers and system integration.
month period. During the period between November and December, Oğuzhan BULUT,
Emrullah ÜLKER in the period between January and February, Oğuz Kaan ULUIŞIK in
the period between March and April, and Utku ÇELİKER in the period between May and
June will be the team leader. The team leader has created major work packages and
determined the distribution of tasks in line with the abilities.

Organization chart

Team Leader
Oğuzhan Bulut

Company
Counselor
Consultant
Emrullah Ülker
Utku Çeliker

Counselor
Oğuz Kaan Uluışık

Power Optimization Electronics and Motion and

Software and
and Energy Saving Structural Design Mechanical Design System Test Leader
Algorithm Leader
Leader Leader Leader Oğuz Kaan Uluışık
Oğuzhan Bulut
Oğuz Kaan Uluışık Utku Çeliker Emrullah Ülker

Power Optimization Electronics and Motion and

Software and System Test
and Energy Saving Structural Design Mechanical Design
Algorithm Engineer Engineer
Engineer Engineer Engineer
Oğuz Kaan Uluışık Emrullah Ülker
Oğuzhan Bulut Oğuzhan Bulut Utku Çeliker

Electronics and
Software and
Structural Design
Algorithm Engineer
Engineer
Utku Çeliker
Emrullah Ülker

6.2 Timeline
In this section, the work to be done during the design process is planned by dividing into
processes and shown on a work time chart. Gann diagram is preferred to show this scheme.
This diagram is shown in Table 1 in the "Annexes" section.
During the project process, this schedule will be followed. The processes completed from
October until the report delivery period are marked in red, and the planned process after
this process is marked in black on the timeline.

6.3 Job Packages

In this section, the work to be done during the design process is planned by dividing into
processes and shown on a work time chart. Gann diagram is preferred to show this scheme.
This diagram is shown in Table 1 in the "Annexes" section.

7. TEST PLAN
In this section, the components of the system to be produced and the tests to be performed
on all
plan will take place. The plans include how and when each test step will take place.
It should include a process to be implemented. Three types of test plans will be made. 1)
Unit testing,
2) Integration test, 3) Acceptance test.

7.1. Unit Tests

In unit tests, the components used on the vehicle and each of which has an important role
in the vehicle's task should be tested separately and checked whether it works as desired. In
other words, the subcomponents of the whole system should be examined unit by unit, the
expected result and the result data should be compared with the test forms, and the errors
should be removed and improved according to the result. The table of unit tests performed
is in the appendices.

7.1.1. Engine Thrust Test

Propellers are attached to the motors and a weight that cannot be lifted and a sensitive scale
is placed one by one. The motors operate at full power respectively and the measured
values are expected to be equal to each other. Finally, the data obtained are compared with
the values given by the manufacturer of the engine.

7.1.2.Raspberry Pi Tests
Camera is integrated into Raspberry Pi device. Display connection is made with an HDMI
cable. The camera hardware is powered up and tested.
GPS is integrated into the Raspberry Pi device. Display connection is made with an HDMI

cable. GPS positions are requested and the results obtained are compared with the current
GPS positions.

7.1.3. Target Recognition and Positioning Test

The object / shape that needs recognition is shown to the camera. Software description of
the object is made. Prepared codes are run and tested. The codes used are given in the
appendices. (ANNEX 1)

7.2. Integration Tests

Integration tests should be done after the unit tests are completed and positive results are
obtained. After the unit tests are completed, the necessary connections should be provided
between these sub-units and larger units should be created. The harmonious and integrated
operation of the sub-units is very important for the vehicle to fulfill its mission completely,
accurately and efficiently. Integration experiments test whether the main modules fulfill
their functions. Test setups should be able to be associated with high-level design, and
should be written by taking into account the characteristics of the design architecture.
7.2.1. Pixhawk Integration with Raspberry Pi
Connection between Raspberry Pi device and Pixhawk 4 control board is made. Display
connection is made with Raspberry pi HDMI cable, then control card integration codes are
entered via python. The codes used and the test table are given in the appendices.
 The drone is made ready for use, and connection tests are performed through "if"
commands. The codes used and the test table are given in the appendices. (ANNEX-3)
(Integration test-2

7.2.3. Target Recognition and Movement Towards Target

After the drone goes to the given GPS position, it is tested to detect the target introduced to
it and move towards it. The codes used are given in the appendices. (ANNEX-4)
(Integration test-2)

7.2.4. Landing and Cargo Drop Test

After the drone goes to the target introduced to it, it is tested to land on the ground and the
mechanical arm to release the cargo. The codes used and the test table are given in the
appendices. (ANNEX-5) (Integration test-3)

7.3. Acceptance Test

After the acceptance test, unit and integration tests are completed with good results, it is
the test determined according to the customer's requirements. The acceptance test is a
formal document that includes the conditions under which the customer will receive the
produced system. Generally, the system is requested to be tested under environmental
conditions of the user. Tests are set up to verify that engineering requirements are met. It
may not be necessary to pass every test for the system to be accepted. The importance of
each test should be stated in the acceptance test document. Since test plans are made at the
beginning of the project, they may need to be reviewed according to the needs arising in
the process. The four basic characteristics of a good requirement are abstraction, precision,
traceability and verifiability. A verifiable requirement provides an objective criterion for
evaluating the system as pass / fail on the test result.
1. https://www.mathworks.com/videos/brushless-dc-motors-introduction-

1564728874059.html

2. https://www.mathworks.com/videos/motor-control-part-3-bldc-speed-control-using-

pwm-1578294719821.html

3. https://www.researchgate.net/publication/335210748_Implementation_of_Autonomou

s_Visual_Detection_Tracking_and_Landing_for_ARDrone_20_Quadcopter

4. https://github.com/tizianofiorenzani/how_do_drones_work/tree/master/scripts

5. https://www.evansville.edu/majors/eecs/downloads/projects2016/CameronRobertsRep

ort.pdf

6. Building An Autonomous Indoor Drone System - eGrove.

7. http://www.set-science.com/manage/uploads/ISAS2019-

ENS_0042/SETSCI_ISAS2019-ENS_0042_0011.pdf.

8. https://www.mathworks.com/videos/image-and-signal-processing-with-matlab-

simulink-82478.html

9. https://stackabuse.com/introduction-to-image-processing-in-python-with-opencv/

10. https://www.raspberrypi.org/documentation/hardware/raspberrypi/bcm2711/rpi_DATA_2711_

1p0_preliminary.pdf

11. https://projects.raspberrypi.org/tr-TR/projects/raspberry-pi-setting-up/5

12. https://github.com/ArduPilot/pymavlink

13. https://dronekit-python.readthedocs.io/en/latest/guide/copter/guided_mode.html

14. https://maker.robotistan.com/raspberry-pi-dersleri-12-dc-motor-kontrolu/
ADDITIONS

Finish Date

November
Start Date

December

February
January
(mouth)
BUSINESS PACKAGES and ACTIVITIES

March
Süre

April

June
May
1. Creating Conceptual Design 20.10.2020 15.12.2020  2
1.1 Literature Review 3
1.2 Conceptual Design 2
2. Detailed Design and Material Selection 15.12.2020 15.02.2021 2
2.1 Detailed Design 2
2.2 Material Selection and Mechanical Design 1
2.3 Updating the Design 1
3. Prototype Manufacturing and System Integration 15.02.2021 10.05.2021 3
3.1 Prototype Mechanical Manufacturing and Assembly 2
3.2 Electric Electronic System Integration 2
3.3 Creating Software and Algorithm 3
4. Creating the Final Report and Planning the Tests 10.05.2021 25.06.2021 1.5
4.1 Unit Tests 1
Communication Test 1
4.2 Integration Tests 1
Systematic Movement and Orientation Tests 1
Load Carrying Tests 1
System Function Tests 1
4.3 Acceptance Test 1
EK-1 EK-2
import cv2 from dronekit import connect, VehicleMode,
from collections import deque LocationGlobalRelative, Command
import numpy as np import time
boyut = 16
from pymavlink import mavutil
ddq = deque(maxlen = boyut)
import argparse
mavi1 = (84, 98, 0)
mavi2 = (179, 255, 255) parser = argparse.ArgumentParser(description='commands')
vid = cv2.VideoCapture(0) parser.add_argument('--connect')
vid.set(3,960) args = parser.parse_args()
vid.set(4,480) connection_string = args.connect
while True: print("Connection to the vehicle on %s"%connection_string)
goo, img1 = vid.read() vehicle = connect(connection_string, wait_ready=True)
if goo:
blur = cv2.GaussianBlur(img1, (11,11), 0)
hsv = cv2.cvtColor(blur, cv2.COLOR_BGR2HSV) EK 3
mask = cv2.inRange(hsv, mavi1, mavi2) Drone yükseklik ve hız ayarları yapılır:
mask = cv2.erode(mask, None, iterations = 2) def arm_and_takeoff(tgt_altitude):
mask = cv2.dilate(mask, None, iterations = 2) print("Arming motors")
(contours,_) = cv2.findContours(mask.copy(), cv2.RETR_EXTERNAL, while not vehicle.is_armable:
cv2.CHAIN_APPROX_SIMPLE) time.sleep(1)
center = None vehicle.mode = VehicleMode("GUIDED")
if len(contours) > 0: vehicle.armed = True
c = max(contours, key = cv2.contourArea) while not vehicle.armed: time.sleep(1)
rect = cv2.minAreaRect(c) print("Takeoff")
((x,y), (width,height), rotation) = rect vehicle.simple_takeoff(tgt_altitude)
s = "x: {}, y: {}, width: {}, height: {}, rotation: #-- wait to reach the target altitude
{}".format(np.round(x),np.round(y),np.round(width),np.round(height),np.round(rotation while True:
)) altitude = vehicle.location.global_relative_frame.alt
print(s) if altitude >= tgt_altitude -1:
rect1 = cv2.boxPoints(rect) print("Altitude reached")
rect1 = np.int64(rect1) break
m = cv2.moments(c) time.sleep(1)
center = (int(m["m10"]/m["m00"]), int(m["m01"]/m["m00"])) arm_and_takeoff(10)
cv2.drawContours(img1, [rect1], 0, (0,0,255),3) #-- set the default speed
cv2.circle(img1, center, 5, (255,255,255),-1) vehicle.airspeed = 7
cv2.putText(img1, s, (25,50), cv2.FONT_HERSHEY_COMPLEX_SMALL, 1, Drone’un gitmesi istenen GPS konumu enlem, boylam ve yükseklik
(255,255,255), 2) cinsinden verilir:
ddq.appendleft(center) print ("go to wp1")
for i in range(1, len(ddq)): wp1 = LocationGlobalRelative(35.9872609, -95.8753037,
if ddq[i-1] is None or ddq[i] is None: continue 10)
cv2.line(img1, ddq[i-1], ddq[i],(0,255,255),1) vehicle.simple_goto(wp1)
cv2.imshow("Orijinal Tespit",img1) Drone hareketleri tanımlama:
if cv2.waitKey(1) & 0xFF == ord("q"): break def set_velocity_body(vehicle, vx, vy, vz):
msg =
vehicle.message_factory.set_position_target_local_ned_encode(
0,
EK-4
0, 0,
import cv2
from collections import deque mavutil.mavlink.MAV_FRAME_BODY_NED,
import numpy as np 0b0000111111000111, #-- BITMASK -> Consider
boyut = 16 only the velocities
ddq = deque(maxlen = boyut) 0, 0, 0, #-- POSITION
mavi1 = (84, 98, 0) vx, vy, vz, #-- VELOCITY
mavi2 = (179, 255, 255) 0, 0, 0, #-- ACCELERATIONS
vid = cv2.VideoCapture(0)
0, 0)
vid.set(3,960)
vid.set(4,480) vehicle.send_mavlink(msg)
while True: vehicle.flush()
goo, img1 = vid.read() if BURASI UYGULAMA SONRASINDA KAMERA
if goo: KONUMUNA GÖRE BELİRLENİP EKLENECEK
blur = cv2.GaussianBlur(img1, (11,11), 0) set_velocity_body(vehicle, gnd_speed, 0, 0)
hsv = cv2.cvtColor(blur, cv2.COLOR_BGR2HSV) elif BURASI UYGULAMA SONRASINDA KAMERA
mask = cv2.inRange(hsv, mavi1, mavi2)
KONUMUNA GÖRE BELİRLENİP EKLENECEK
mask = cv2.erode(mask, None, iterations = 2)
mask = cv2.dilate(mask, None, iterations = 2) set_velocity_body(vehicle,-gnd_speed, 0, 0)
(contours,_) = cv2.findContours(mask.copy(), cv2.RETR_EXTERNAL, elif BURASI UYGULAMA SONRASINDA KAMERA
cv2.CHAIN_APPROX_SIMPLE) KONUMUNA GÖRE BELİRLENİP EKLENECEK
center = None set_velocity_body(vehicle, 0, -gnd_speed, 0)
if len(contours) > 0: elif BURASI UYGULAMA SONRASINDA KAMERA
c = max(contours, key = cv2.contourArea) KONUMUNA GÖRE BELİRLENİP EKLENECEK
rect = cv2.minAreaRect(c)
set_velocity_body(vehicle, 0, gnd_speed, 0)
((x,y), (width,height), rotation) = rect
s = "x: {}, y: {}, width: {}, height: {}, rotation: Drone başlangıç noktasına geri dönüş:
{}".format(np.round(x),np.round(y),np.round(width),np.round(height),np.round(rotation)) vehicle.mode = VehicleMode("RTL")
print(s)
rect1 = cv2.boxPoints(rect)
rect1 = np.int64(rect1)
m = cv2.moments(c)
center = (int(m["m10"]/m["m00"]), int(m["m01"]/m["m00"]))
cv2.drawContours(img1, [rect1], 0, (0,0,255),3)
cv2.circle(img1, center, 5, (255,255,255),-1)
cv2.putText(img1, s, (25,50), cv2.FONT_HERSHEY_COMPLEX_SMALL, 1,
(255,255,255), 2)
ddq.appendleft(center)
for i in range(1, len(ddq)):
if ddq[i-1] is None or ddq[i] is None: continue
cv2.line(img1, ddq[i-1], ddq[i],(0,255,255),1)
cv2.imshow("Orijinal Tespit",img1)
if cv2.waitKey(1) & 0xFF == ord("q"): break
 FCB (Continued)
EK-5 # WAYPOINT
arm_and_takeoff(1) komut.add(Command(0, 0, 0,
import RPi.GPIO as GPIO mavutil.mavlink.MAV_FRAME_GLOBAL_RELATIV
FCB Code from time import sleep E_ALT,
GPIO.setmode(GPIO.BOARD) mavutil.mavlink.MAV_CMD_NAV_WAYPOINT, 0,
Motor1A
from dronekit import= Command,
16 connect, 0, 0, 0, 0, 0, -35.36265286, 149.16514170, 20))
VehicleMode,Motor1B = 18
LocationGlobalRelative komut.add(Command(0, 0, 0,
Motor1E = 22 mavutil.mavlink.MAV_FRAME_GLOBAL_RELATIV
import time
GPIO.setup(Motor1A,GPIO.OUT) E_ALT,
from pymavlink import mavutil
GPIO.setup(Motor1B,GPIO.OUT) mavutil.mavlink.MAV_CMD_NAV_TAKEOFF, 0, 0,
iha = connect("127.0.0.1:14550",
GPIO.setup(Motor1E,GPIO.OUT) 0, 0, 0, 0, 0, 0, 1))
wait_ready=True)
Print“Ileri hareket”
def takeoff(irtifa): komut.add(Command(0, 0, 0,
GPIO.output(Motor1A,GPIO.HIGH)
while iha.is_armable is not True: mavutil.mavlink.MAV_FRAME_GLOBAL_RELATIV
GPIO.output(Motor1B,GPIO.LOW)
print("İHA GPIO.output(Motor1E,GPIO.HIGH)
arm edilebilir durumda değil.") E_ALT,
time.sleep(1) sleep(2) mavutil.mavlink.MAV_CMD_NAV_TAKEOFF, 0, 0,
print("İHA Print“Gerihareket”
arm edilebilir.") 0, 0, 0, 0, 0, 0, 10))
iha.mode =GPIO.output(Motor1A,GPIO.LOW)
VehicleMode("GUIDED") # RTL
iha.armedGPIO.output(Motor1B,GPIO.HIGH)
= True komut.add(Command(0, 0, 0,
GPIO.output(Motor1E,GPIO.HIGH)
while iha.armed is not True: mavutil.mavlink.MAV_FRAME_GLOBAL_RELATIV
print("İHA sleep(2)
arm ediliyor...") E_ALT,
print “Motor durdu”
time.sleep(0.5) mavutil.mavlink.MAV_CMD_NAV_RETURN_TO_L
print("İHA GPIO.output(Motor1E,GPIO.LOW)
arm edildi.") AUNCH, 0, 0, 0, 0, 0, 0, 0, 0, 0))
GPIO.cleanup()
iha.simple_takeoff(irtifa) # DOĞRULAMA
while iha.location.global_relative_frame.alt < komut.add(Command(0, 0, 0,
irtifa * 0.9: mavutil.mavlink.MAV_FRAME_GLOBAL_RELATIV
print("İha hedefe yükseliyor.") E_ALT,
time.sleep(1) mavutil.mavlink.MAV_CMD_NAV_RETURN_TO_L
def gorev_ekle(): AUNCH, 0, 0, 0, 0, 0, 0, 0, 0, 0))
global komut komut.upload()
komut = iha.commands print("Komutlar yükleniyor...") Test
Test Plan Name
komut.clear()
Drone Unit Tests takeoff(10) Iden 5.1.
time.sleep(1) gorev_ekle() tific
# TAKEOFF komut.next = 0 atio
komut.add(Command(0, 0, 0, iha.mode = VehicleMode("AUTO")n
while True:
mavutil.mavlink.MAV_FRAME_GLOBAL_RELATIV Num
E_ALT, next_waypoint = komut.next
ber:
print(f"Sıradaki komut {next_waypoint}")
mavutil.mavlink.MAV_CMD_NAV_TAKEOFF, 0, 0,
0, 0, 0, 0, 0, 0, 10)) Unit tests and results of all sub-elements
time.sleep(1)in the WhiteBox
if next_waypoint is 4: ☒
cargo drone system
Definition print("Görev bitti.") Type Black box
break
print("Döngüden çıkıldı.")
☐
Experiment Performer
Project team Date

Hardware Version Hour

Information on the setup Cargo Drone subunits

No Step

Action Expected Result Comments

UD
G

Engines to be able to Engines run at full power and tested at

Efficient handle the total weight maximum weight.
01
operation of
✓
engines
Raspberyy pi
Raspberry Pi Raspberry installation has been
installation
02 operating
should result in
✓ successfully completed. Control card,
camera and GPS modules can be
success
added.

The camera
Camera-GPS
03
module
should be able to ✓ Camera operation has been observed and
be turned on and GPS coordinates have been taken at
operation
off and the GPS expected values..
data should give
accurate data.

40
 Code written in the The code worked efficiently and
Target
04
recognition
Python program ✓ detected the object identified as a blue
should work cross from different angles and
software
without errors . distances.
Overall Results ✓

P: Pass; F: Fail; CA: Could not apply

Test
Test Plan Name Integration Test-2 Iden 5.2.2. ve 5.2.3.
tifica
tion
Num
ber:

The drone is made ready for flight. Display White Box

Definition connection is provided to Raspberry pi device Type ☒
via HDMI cable. The purpose of this test is to Black Box
test the Cargo Drone's application of the ☐
commands given to it in the python interface
and to verify the data with the screen output.
Experiment
Project team Date
Performer
Hardware Time
Version
Step Np

UD

Action Expected Result Comments

G

Making the Successful take- The drone has correctly performed the
0
drone ready off and landing
✓ ascent / descent commands given to it in the
1
for flight of the drone Python interface. With the help of the meter,
its height was measured and verified with the
desired data.
Direction Correct East / West and North / South movements are
0
al recognition of
✓ defined as X (+ -) and Y (+ -). Their
2
moveme compass movements in these directions were tested and
nts directions and their controls were matched with the compass.
movement
Moveme Accessing the It has completed the movement to the GPS
0
nt to given GPS
✓ location entered. The distance difference is
3
given position compared with the measured GPS data sheet
location data. The margin of error is due to GPS. The
by GPS system is working properly.
Fixing Returning from Memorized the position of the starting point.
0
the the shortest
✓ The position received by the print function
4
starting route to the coincided with the origin position. The return
point and starting point command in the resource library has returned
return assigned as a to the starting point with a tested success.
GPS code
Recognizing the The camera detected the place defined by the
0
Image place
✓ cross by entering the field of view. The drone
5
detection determined with made the determination of the movement in a
and the cross and way to take the vertical position of the cross.
moveme coming to the
nt to the region so that

41
 area the route
determination is
perpendicular
Overall Results ✓
P: Pass; F: Fail; CA: Could not apply

Test
Test Plan Name Integration Test-1 Identifi 5.2.1.
cation
Numbe
r:

Includes integration and operation of

Raspberry Pi and Pixhawk-4 control board Type White Box
Definition ☒
Black box ☐

Experiment
Project team Date
Performer

Hardware Version Time

Step No

Action Expected Result Comments

UD
G

Integration of Raspberry Raspberry Pi device is connected to Pixhawk

0
control cards and Pixhawk
✓ control board. Pixhawk-4 driver has been
1
must be able installed on the Raspberry device and the
to receive / connection of the control card hardware is
send signals confirmed on the screen connected for control
to each other. purposes.
Executing Ability to The outputs of the test codes written in Python
0
Python perform
✓ were checked. Confirmation of the printouts
2
tasks software was requested using the If / Print commands.
functions The expected results have been obtained.
Overall ✓
Results
P: Pass; F: Fail; CA: Could not apply

42
 Test
Integration Test-3 Iden 5.2.4.
Test Plan Name tifica
tion
Num
ber:

Test of the drone that reaches the cargo drop White Box
Definition area by means of a mechanical arm. Type ☒
Black box
☐
Experiment
Project team Date
Performer
Hardware Version Time
Adım No

Action Expected Result

UD

Comments
G

Descent The drone The drone verifies its location in the Python
0
should verify
✓ interface both in the GPS interface and via the
1
through its camera. With the outputs being "true", it made
software that it the descending movement in a controlled
has reached the manner and reached the desired height level..
coordination
given to it and
begin its landing
motion.
Cargo Cargo arm The drone positioned perpendicular to the
0
handle taking /
✓ cargo on the ground has successfully
2
opening releasing the received / released the cargo.
and 5x5x5 cm load
closing given to it
Overall Results ✓
P: Pass; F: Fail; CA: Could not apply

43
GAZI IS THE FUTURE…

44
45