FYP Research Paper SSUET FINAL
FYP Research Paper SSUET FINAL
FYP Research Paper SSUET FINAL
Abstract—Through utilizing the modern discussion on product delivery hexacopter design that
telecommunication technology, this thesis presents method will be used for autonomous control. It is difficult to
for building an autonomous hexacopter which is an design a big hexacopter and the ability to lift heavy
unmanned aerial vehicle with six rotors. It is capable of weight. In this paper, the purpose is the ability to fly and
carry product of certain weight and deliver it to desired
lift certain loads so that design becomes simpler and
destination. Through the use of Ardupilot software, users
are able to specify remote waypoints for the hexacopter to compatible. Hexacopter with heavy payload has also
navigate. The hexacopter is controlled through an been researched in [3] but it still in the design stage and
ArduPilot, employing GPS modules, along with a suite of has not flown yet while the power delivered by the
sensors including a camera. Utilizing various image tethered cables that limit its flight altitude. The other
processing algorithms for object detection, coordinates are researcher discussed design and construction hexacopter
assigned to the hexacopter which generate flight pattern so for high endurance [4].
that hexacopter perform a task. With the help of GPS
module and image processing techniques, the hexacopter I.II. Motivation Of Project:
can perform an intelligent searching and delivering All the previous work based on complicated
operations. calculations and algorithms. However, in this project
Keywords—Unmanned aerial vehicle (UAV), autonomous
hexacopter configuration would carry stuff of certain
drone, Raspberry pi, navigation system based drone, Open-CV, weight and reaches the given location and drop that stuff.
python, payload, thurst However, before dropping the product, the camera
detects the person to whom the product may deliver. Face
I. INTRODUCTION detection algorithm is used to do that task.
Unmanned aerial vehicles (UAVs) are aircraft with In designing the hexacopter, it is necessary to consider
no on-board crew or passengers. They can be automated the following factors: a total weight that can be lifted, the
‘drones’ or Remotely piloted vehicles (RPVs). UAV’s thrust, motor type, propeller type and size, the material
can fly for long periods of time at a controlled level of composition of the frame, and flight time. Due to these
speed and height and have a role in many aspects of factors, it is expected that the design could achieve the
aviation. A multirotor is a type of aircraft which is objectives and overcome flight performance problems
resemblance to the helicopter but have more than two and the change of the load distribution due to the
rotors. The most common amount of rotors are 3 additional load [5]. The next step is to customize the
(tricopter), 4 (quadcopter), 6 (hexacopter) and 8 design by supporting components such as Electronic
(octacopter), however any configuration is possible Speed Controller (ESC), Brushless DC motor, battery,
according to the requirement of task. Multirotors have flight controller, GPS and compass and remote controller.
attracted a lot of attention for research in recent years The entire operation would be done as much as simpler
due to their maneuverability, simple construction, than before.
flexibility and ability to take a payload. Today their main
commercial use is related to aerial photography, II. OBJECTIVES
surveillance and remote sensing.
Objectives of the project are as follow:
I.I. Previous Work: i. Aim to design hexacopter capable of delivering
The previous work, heavy-lift hexacopter control has product of certain weight to the specified
been used neural network[1, 2] but it discussed about destination autonomously.
movements of hexacopter and control system that used ii. Usage of Simple Open-CV algorithm for face
in hexacopter. While in this paper the focus of detection of a person to whom product deliver.
SSUET Proceedings for FYPEE Batch 2016 @ 2019
iii. Navigation system interface with Google Map.
III. METHODOLOGY C. Flow chart of the methodology
A. Background
Unmanned Aerial Vehicle (UAV) becomes popular
in the world of science because of its various
applications in this life such as search and rescue,
monitoring, firefighting, surveillance, agriculture, aerial
photography and others. This project is based on the
lifting stuff and deliver it another specified location
autonomously.
In the first place, select the components and body
frame of hexa-copter. Assemble the drone by connecting
ESCs and motors and Li-Po battery with body frame.
Calibrate the RPM of motors to stabilize required thrust
as well as weight of drone. Calibrate the APM and set
the coordinate through software. Test first flight without
load and then gradually increase load to check the
theoretical weight to be lifted. Now test an autonomous Fig. 2. Flow diagram of project
flight by setting the coordinates of destination. For face
detection and gripper movement, upload a program to
the memory of Raspberry-Pi. Installation of Raspberry- The sequence of working of project is described in the
Pi on the drone body. Finally, an autonomous flight Fig. 2. At the base station the assignment of coordinates
along the package to deliver at the destination. of the specified destination is given to the hexacopter’s
apm. An autonomous flight of drone is started from the
B. System architecture base station towards the destination. If any failure
happens to drone during the flight like any rotor stops
Project consist of numbers of sensors and working or low battery indication, then it get back to the
equipment in order to achieve the objective. base station. In case of no failure, the flight proceeding
Raspberry pi, sonar and gripper is connected with the towards destination. When the drone is reached at
raspberry pi. Sonar is used to detect the obsticle assigned location then camera mode is activated and
during the flight which may help the drone to avoid capture the image of a person at a particular height and
collision. However, camera is used to capture the detect the face. If no face detection occurs due to any
image and detect the face of a person. Gripper is used unfortunate situation, the drone gets back to base station
to carry the stuff to be delivered. Raspberry pi is again. If face is detected, then drone drops the product
connected with apm. Apm contain built-in through gripper and fly back to the base station.
accelerometer, gyroscope and proximity sensors
which helps the drone to perform a successful flight. D. Components of the project
Drone is controlled through RC transmitter. The In order to build such kind of prototype, following
entire configuration is shown in Fig. 1. components are mandatory:
BLDC motors: Rotors or motors are responsible to Telemetery: 3DR RADIO is the ideal way to setup
produce thrust in order to fly the drone. Selection of telemetry connection between APM and a ground
motors should be based on efficiency, torque and RPM. station. It gives the real time data of drone to the
Six Turningy typhoon 500H heli motors are used to ground station. It uses an open-source firmware which
build this project. Each motor has 1800 rpm per volt allows to do things that cannot be done with other
with the maximum current rating of 55 amperes and also radios.
has thrust of about 6 kilograms.
APM 2.8: Ardupilot is an on-board circuit used for drone Fig. 8. HC-SRO4 Sonar
flight stabilization. It has two ways telemetry with
wireless module and built-in compass, accelerometer, Raspberry pi: A single 32 bits or 64 bits on-board
gyroscope, proximity sensors and others essential computer. It has an operating frequency of 700 MHz and
sensors. 512 MB ram. It is powerful and used to perform various
SSUET Proceedings for FYPEE Batch 2016 @ 2019
task through programming. In this project raspberry pi
module B is used for face detection.
C. Equations V. CONCLUSION
TABLE I: COMPONENTS WEIGHT An autonomous hexacopter product delivery is an
No Item Unit Mass (gr)
innovative approach for industries to save fuel
consumption as well as time. However, this project has
1 Battery 1 6016 certain features but with less complication. This project
2 Motor 6 1020 could be used in smart cities as well.
3 ESC 6 300
4 Propellers 6 4800
5 Body 1 2760
ACKNOWLEDGMENT
6 APM 1 43 We would like to acknowledge the electronic
engineering department, SSUET for giving us the
7 GPS 1 20.6
opportunity to present and complete the project. Besides,
8 3DR telemtery 1 19.4 also gratitude to the internal advisor, all lab technician,
TOTAL 14,979.0 lecturer and FYP committee for their utmost guidance
and support throughout the project.
With a total weight of 14,979.0 grams is known then
the thrust of each motor is determined as follows
REFERENCES