VISVESVARAYA TECHNOLOGICAL UNIVERSITY

BELAGAVI-590018

Project Report
On
(18CSP82)

“Drone Delivery with Object Detection”

Submitted in the partial fulfilment of the requirement for the award of the degree in

Bachelor of Engineering
In
Information Science & Engineering
Submitted by
Ms. M. LIKHITHA SHREE 1NC18IS026
Ms. M. SUMA RANI 1NC18IS027
Mr. NIROOP S RAO 1NC18IS033
Mr. ANANYA VIKRAM 1NC18IS026

Under the Guidance of

Mrs. SANDHYA L
Asst. Professor, Dept. of ISE

DEPARTMENT OF INFORMATION SCIENCE AND ENGINEERING

NAGARJUNA COLLEGE OF ENGINEERING AND TECHNOLOGY

(An Autonomous Institution under VTU, Accredited by NAAC with “A+” Grade)
BENGALURU– 562164

2021-2022
NAGARJUNA COLLEGE OF ENGINEERING AND TECHNOLOGY
(An Autonomous Institution under VTU, Accredited by NAAC with “A+” Grade)

Bengaluru-562164, Karnataka, India

DEPARTMENT OF INFORMATION SCIENCE & ENGINEERING

CERTIFICATE

Certified that the project work entitled “DRONE DELIVERY WITH OBJECT DETECTION”
carried out by Ms. M. LIKHITHA SHREE (1NC18IS026), Ms. M. SUMA RANI (1NC18IS027), Mr.
NIROOP S RAO (1NC18IS033) , Mr. ANANYA VIKRAM (1NC17IS003) bonafide students of
Nagarjuna College of Engineering and Technology, an autonomous institution under Visvesvaraya
Technological University, Belagavi in partial fulfilment for the completion of course Project evaluation
for the award of Bachelor of Engineering in Information Science & Engineering during the academic
year 2021-2022. It is certified that all corrections/suggestions indicated for internal assessment have been
incorporated in the report deposited in the departmental library. The project work has been approved, as it
satisfies the academic requirement in respect of project work prescribed for the said degree.

Name & Signature of the Guide Name & Signature of the HOD Name & Signature of the Principal
Mrs. Sandhya L Dr. Anil Kannur Dr. B. V Ravishankar
Asst. Professor, Dept of ISE HOD, Dept of ISE Principal, NCET

External Viva-Voice

Name of the Examiner Signature with date

1.......................................... ..........................................

2.......................................... ..........................................
ACKNOWLEDGEMENT

Every project begins with an idea and materializes with concrete efforts. In the beginning, we would
like to thank the almighty God and our parents who gave us the strength and capability to work on this
project and complete it successfully.

We are extremely grateful to our project guide Mrs. Sandhya L, Assistant Professor, Department of
Information Science & Engineering for the guidance and encouragement.

It is indeed gratifying to have the privilege to express our sense of gratitude to our project coordinator
Ms. Shruti Jalapur, Assistant Professor Department of Information Science and Engineering, NCET
for her scholarly guidance during the course of investigation.

We extend our sincere gratitude to Dr. Anil Kannur, Professor & Head of the Department,
Information Science & Engineering, NCET, for his consistent assistance and guidance during the course
of the project work.

We also express our gratitude to Dr. B V Ravishankar, Principal, Nagarjuna College of Engineering
and Technology for his help and support.

Finally, we express our immense pleasure and thanks to all Teaching staff and non-teaching staff of
the Department of Information Science & Engineering, NCET for their co-operation and support.

Mr. ANANYA VIKRAM 1NC17IS003

Ms. M. LIKHITHA SHREE 1NC18IS026

Ms. M. SUMA RANI 1NC18IS027

Mr. NIROOOP S RAO 1NC18IS033

i
ABSTRACT

Unmanned aerial vehicles (UAVs), or drones, have been successfully applied to disaster management, 3D
mapping and precision agriculture. A delivery drone is an unmanned aerial vehicle (UAV) used
to transport packages, medical supplies, food, or other goods. Delivery drones are typically autonomous.
Drones can be used to transport medicinal products such as blood products, vaccines, pharmaceuticals.
And, for rapid delivery of prepared foods, such as pizzas, tacos, and frozen beverages. Recently, drones are
considered as a promising solution to rapid parcel delivery. Drones will soon be delivering the packages,
but the concerns are receiving damaged packages, interference from people and weather, and packages
being delivered to the wrong address. This project is mainly concerned with safe and secure delivery.

The main objective of our model is to detect the object(product for delivery) and notify both sender and
receiver the details of the object. This variant uses a PIR infrared sensor, which is installed in the delivery
cabinet, to find the object. When an object is placed, the sensor notices it and sends a message to the end
users along with a picture of the object with the use of esp32 cam and sim 800l module.

This drone can be very helpful in emergency situations and in places where there are still gaps in the
availability of reliable transportation services. In any pandemic situation where, human interaction is not
advised and social distancing is the key in health management structure, this Unmanned drone
delivery system can be used as a strong weapon to fight against the pandemic.

ii
PARTICIPATION

Our Project was sponsored by Karnataka State Council of Science & Technology(KSCST). KSCST is a
council of Indian Institute of Science in Karnataka. KSCTC chooses few projects from all the college
students in Karnataka and sponsor a specific amount after selection every year. We had been sponsored
Rs.6,000 under 45th series SPP. The Midterm evaluation of our project was held on 28th June 2022 through
Google Meet. The Final evaluation of our project will be in July-August.

iii
TABLE OF CONTENTS

CHAPTER NO. CHAPTER NAME PAGE NO

ACKNOWLEDGEMENT i
ABSTRACT ii
PARTICIPATION iii
TABLE OF CONTENTS iv
LIST OF FIGURES v
LIST OF ABBREVIATIONS vi

CHAPTER 1 INTRODUCTION 1
1.1 Overview 1
1.2 Introduction To Drone Delivery 2
1.3 Evolution in Delivery Drones 2
1.3.1 Why Delivery Drones 4
1.4 Problem Statement 5
CHAPTER 2 LITERATURE SURVEY 6
CHAPTER 3 SYSTEM REQUIREMENTS AND SPECIFICATIONS 9
3.1 Hardware Requirements 9
3.2 Drone Components And Mechanism 10

CHAPTER 4 SYSTEM ANALYSIS 12

4.1 Purpose Of Research 12

4.2 Advantages And Applications 12

4.3 Obstacles To Overcome By Drone Delivery 13

CHAPTER 5 SYSTEM DESIGN 14

5.1 Methodology 14

5.2 Design And Workflow 14

CHAPTER 6 IMPLEMENTATION 18

6.1 Model Components And Working 18

6.2 Introduction to ESP32-CAM 19

6.3 Introduction to Arduino UNO 20

6.4 Introduction to PIR Sensor 20

6.5 Modules 22

6.5.1 Object Detection Module 22

6.5.2 Message Alert Module 22

6.5.3 Delivery Module 23

CHAPTER 7 RESULT ANALYSIS AND DISCUSSION 24

CHAPTER 8 CONCLUSION AND FUTURE SCOPE 28

References 29
LIST OF FIGURES

FIGURE NO. FIGURE NAME PAGE NO.

Figure 5.2.1 Flowchart of working model 15

Figure 5.2.2 Flowchart of extracting user’s location 16

Figure 6.1 Components of Drone Model 18

Figure 6.2 ESP32-CAM Module 19

Figure 6.3 Arduino UNO board 20

Figure 6.4 PIR Sensor 21

Figure 6.5.1 ESP32 CAM to Arduino UNO board 22

Figure 6.5.2 Arduino UNO board to GSM SIM800L 22

Figure 6.5.3 Working procedure of delivery module 23

Figure 7.1 Working model of drone 25

Figure 7.2 Drone Cabinet 25

Figure 7.3 Product for delivery 26

Figure 7.4 Connections on UNO board 26

Figure 7.5 Bot Testing 27

LIST OF ABBREVIATIONS

Sl. No. Short form Abbreviation

1. FPV First Person View

2. QC Quadcopter

3. UAV Unmanned Aerial Vehicle

4. GSM Global System for Mobile Communication

5. ESC Electronic Speed Controller

6. PIR Passive Infrared Sensor

7. GPRS General Packet radio service

8. FC Flight Controller

9. LOS Line Of Sight

10. GPIO General Purpose Input/Output

Drone Delivery with Object Detection

CHAPTER-1
INTRODUCTION

1.1 Overview

The technological evolution that characterized the last century profoundly changed
the way people perform daily activities. This is evident when considering, for instance,
the radical transformations brought in the organization of house works by
refrigerators/freezers, washing machines, etc. Among such innovative solutions,
particular attention has been dedicated to “drones” (unmanned aircrafts or ships guided
by remote control or onboard computers), and to Unmanned Aerial Vehicles (UAVs,
aircrafts piloted by remote control), which recently became widely available in the
commercial marketplace. In fact, apart from merely recreational applications in which
commercial drones and UAVs are used in a way similar to a radio-controlled car, they
could represent a quite cheap instrument that allows users to access areas that are
normally difficult/dangerous to reach.

Another field in which drones and UAVs could soon bring enormous changes is
represented by delivery services, as it has been shown by prototypes developed by
Amazon, Google and DHL. In particular, Amazon is waiting for the permission from the
U.S. Federal Aviation Administration in order to use its drones to carry out deliveries in
less than 30 minutes. Similarly, in the summer of 2014, Google tested the UAV-based
delivery of first-aid kits in Australia, by replicating the successful experience of DHL
dated December 2013, in which a “parcelcopter” was able to deliver urgent medications
to customers.

As a consequence, the use of drones in last mile delivery introduces new issues to
handle. The future challenge underlying this application is not so much the design of
drones for parcel delivery, but to provide the security to customer (i.e., sender/receiver).
In our model we focus on detecting the object and notifying the customers with the details
of the delivering object.

DEPT. OF ISE, NCET, BENGALURU 2021-2022 1

Drone Delivery with Object Detection

1.2 Introduction To Drone Delivery

A delivery drone is a type of unmanned aerial vehicle (UAV) used for distributing
packages to consumers during the last mile delivery process. These types of drones
generally have 4-8 propellers, rechargeable batteries and the ability to carry lightweight
containers. They can be operated either autonomously using AI technology or remotely,
with distribution centres and operators overseeing the flight. With drone technology
already advancing, the capability of avoiding collisions with other drones in flight,
navigating through urban landscapes and employing smart landing techniques already
exists.

Benefits of drone delivery are currently being tested but could include lower costs,
higher operational efficiency, new revenue streams, instantaneous fulfilment, less
congested roadways, fewer accidents and lower emissions.

As e-commerce continues to grow and traditional forms of delivery are no longer

the most efficient option, delivery companies are experimenting with the implementation
of drones. The current most popular use cases for delivery drones are time-sensitive
materials such as medicine and food or small items for same-day delivery.

1.3 Evolution In Delivery Drones

In February 2014, the prime minister and cabinet affairs minister of the United
Arab Emirates (UAE) announced that the UAE planned to launch a fleet of UAVs for
civilian purposes. Plans were for the UAVs to use fingerprint and eye-recognition
systems to deliver official documents such as passports, ID cards and licenses, and supply
emergency services at accidents. A battery-powered prototype four-rotor UAV about half
a meter across was displayed in Dubai.

The concept of drone delivery entered the mainstream with Amazon Prime
Air – Amazon.com founder Jeff Bezos' December 2013 announcement that Amazon was
planning rapid delivery of lightweight commercial products using UAVs. Amazon's press
release was met with scepticism, with perceived hurdles including federal and state
regulatory approval, public safety, reliability, individual privacy, operator training and

DEPT. OF ISE, NCET, BENGALURU 2021-2022 2

Drone Delivery with Object Detection

certification, security (hacking), payload thievery, and logistical challenges.

In December 2013, in a research project of Deutsche Post AG subsidiary DHL, a

sub-kilogram quantity of medicine was delivered via a prototype Microdrones
"Parcelcopter", raising speculation that disaster relief may be the first industry the
company will use the technology.

In July 2014 it was revealed Amazon was working on its 8th and 9th drone
prototypes where each could fly 50 mph (80 km/h) and carry a 5 lb (2.3 kg) package, and
had applied to the FAA to test them.

In August 2014, Google revealed it had been testing UAVs in Australia for two
years. The Google X program known as "Project Wing" announced an aim to produce
drones that can deliver products sold via e-commerce.

In September 2014, FedEx was reportedly testing integration of drone delivery

with their existing logistics model.

In February 2015, Hangzhou-based e-commerce provider Ali Baba started

delivery drone service in a partnership with Shanghai YTO Express in which it delivered
tea to 450 customers around select cities in China.

In 2015, an Israeli start-up Flytrex partnered with AHA,[34] Iceland's

largest eCommerce website, and together they initiated a drone delivery route which
demonstrated reducing delivery time from 30 minutes, to less than 5 minutes.

In March 2016, Flirtey conducted the first fully autonomous FAA approved drone
delivery in an urban setting in the U.S.

In April 2016, a joint project in Japan involving the central government, Chiba
City, research institutions and companies including Rakuten was launched to trial home
drone deliveries in an urban area. A similar test project was carried out in Naka,
Tokushima in February 2016 as a way to facilitate shopping for people who live in a
depopulated area.

In December 2016, Amazon Prime Air made its first delivery using a drone in the
United Kingdom.

DEPT. OF ISE, NCET, BENGALURU 2021-2022 3

Drone Delivery with Object Detection

In China, JD.com has been developing drone delivery capabilities. As of June

2017, JD.com had seven different types of delivery drones in testing across four provinces
in China (Beijing, Sichuan, Shaanxi and Jiangsu). The drones are capable of delivering
packages weighing between 5 and 30 kg (11 to 66 lbs) while flying up to 100 km/hr
(62 mph).

In January 2018, Boeing unveiled a prototype of a cargo drone for up to 500 lb

(227 kg) payloads, an electric flying testbed that completed flight tests at the Boeing
Research & Technology research centre in Missouri.

The drone industry is expected to command a staggering 90 billion dollar market

by 2025. so, you can imagine the drone delivery industry in the near future due to the
great solutions that can be presented through.

Companies like: Amazon, Google, Wal-Mart, FedEx and UPS are all seriously
investing in drones’ technology.

For many logistics providers, it’s clear that if drone technology can be properly
regulated . the opportunities to cut costs and speed transportation are huge.

1.3 .1 Why Delivery Drones?

Here are just some of the reasons why the business world is excited, by the
possibilities of drone delivery:
1. Resolving the Last Mile by drone delivery
Last mile of delivery between the warehouse and customer, is often the most cost-
intensive. Planning labour requirements and routes to ensure as many packages arrive at
their destinations, as quickly as possible can be a logistical nightmare.
Drones can ensure each delivery, is carried out safely and quickly to the correct address,
securing high levels of customer satisfaction even as costs are reduced.
2. Improving the Supply Chain by drone delivery
Drones are being used to deliver raw materials from mining sites to processors,
such as:
• Camera equipped drones are being used for surveillance in company warehouses.
• Other machines are being used to aid maintenance.

DEPT. OF ISE, NCET, BENGALURU 2021-2022 4

Drone Delivery with Object Detection

• Repair efforts by transporting tools, and parts to remote factory sites or off- shore
rigs.
• Over the next 5 years, the use of drones for inspection, delivery, surveillance and
basic tasks is expected to explode.

1.4 Problem Statement

Drones will soon be delivering the packages, but are consumers really ready for
what this will mean for their privacy?

Other concerns people have are receiving damaged packages, interference from
people and weather, and packages being delivered to the wrong address.

This project is mainly concerned with safe and secure delivery.

DEPT. OF ISE, NCET, BENGALURU 2021-2022 5

Drone Delivery with Object Detection

CHAPTER-2
LITERATURE SURVEY

We conducted our literature review in several steps. We chose the databases to

search in based on available access which led to following databases:

• IEEE Xplore
• Research Gate
• JSCDSS

The paper [1] discusses the potential value of introducing modularity to a drone
delivery system. The main reason to introduce modular architecture is the limited battery
capacity in drones. The results in this paper show that the drone delivery system is
unstable with the static management strategy for both modular and non-modular drone
operations. Although the drone delivery system is stable with both dynamic programming
and the proposed method, the latter can save a delivery time and energy. Also, the results
show that modular drone delivery system has potential to improve delivery time
compared to a non-modular system. In this paper, a fully modular system is compared
with a non-modular drone system. The results indicate that a semi-modular system with
a replaceable battery is an interesting case to analyse for future work.

In the paper [2] DDAM concept was introduced, examined and evaluated by
experts through the Design Science Research Guideline (DSRG). Through the seven
guidelines, this paper gathered various opinions on DDAM as well as valuable insights
to future last-mile delivery and traffic problems. First, using drones could increase the
speed and flexibility of a delivery process, eventually enlarging delivery capacity.
Second, autonomous mobility takes place as a middle process of a delivery, which reduces
vehicles by adding value to existing transportation methods. However, there remains
limitations for concept development, providing potential further research areas. First, the
concept evaluation was conducted by only five evaluators from two companies, which
limited the research from gathering wider point-of-views. The evaluations were narrowly
focused in the perspective of each expert’s expertise, which enabled critical views from
various focus areas but restricted the experts from evaluating the concept’s overall idea.

DEPT. OF ISE, NCET, BENGALURU 2021-2022 6

Drone Delivery with Object Detection

Second, DDAM is a combination of two futuristic features, which are not yet
commercialized.

Thus, the concept understanding of each expert could have varied, depending on
personal insights to the futuristic features. Lastly, the concept was not numerically
compared to any existing concepts. Limitations were found to be, this model uses road
way as the path, which takes longer time. There’s a need of GPS location on both the
sides (start point to destination), till the delivery is completed.

The paper [3] focuses on the development of prototype system that is able to fly to
the approximate delivery location using GPS and then find the exact drop-off location
using visual navigation. This system is based on the open source project ROS. The
collision avoidance only uses data from the front facing camera, this can be problematic
especially when the drone is navigating in tight spaces or is trying to land. Adding an
additional bottom-facing (depth) camera could alleviate this issue. Currently the drone is
solely relying on visual data for navigation and stabilization. Combining the camera
images with data from the IMU could result in more stable flight trajectories and thus
allowing for faster flight speeds. The limitations are it uses an autonomous drone with a
front facing camera, this can be problematic especially when the drone is navigating in
tight spaces or is trying to land.

In the paper [4] Quality control of the RBC solution for transfusion transported to
a remote island using a drone was achieved. The cross-matching tests were unremarkable.
One of the reasons for the success of this trial was that the weather had little effect on the
experiment Temperature is the most important factor affecting the quality of RBC
solutions during transportation. The UH transport was able to transport blood under
controlled temperatures, and the actual analysis of the blood showed that it was possible
to meet the rapid demand for blood transfusions anywhere in Japan, including remote
areas. Blood transported at an appropriate temperature can also be used before the
expiration date if the blood was not used upon arrival. This trial is the first to transport
packs of RBC solution for transfusion beyond a part of the western Pacific Ocean. But
Safety and Security were not considered.

In Paper [5] Drone Delivery system will track live location of consumer and hence
provide assured delivery of package to the correct place and correct person. GPS Drone

DEPT. OF ISE, NCET, BENGALURU 2021-2022 7

Drone Delivery with Object Detection

delivery system will locate the consumer through GPS and detect the live location and
deliver the package accurately and within a stipulated time as per the date of delivery.
GPS will be inbuilt in the drone and live tracking device will detect the location of user’s
device which will currently be handled by the customer using OTP provided during
generation of package to avoid fraud.

There will be a main warehouse and some sub-warehouses, the drone will pick up
the package from the nearest warehouse of the delivery location. It will behave like any
home delivery app where both the customer and the dispatcher will be able to trace the
live location of the package. These drones will ensure the correct delivery of their
demands in their current positions that is wherever they are.

The report [6] examines the value chain and opportunities in the delivery drone’s
market. It also discusses the barriers for adoption. It concludes with our case for drones
to handle the last mile of delivery of most lightweight packages. This report represents
Quadcopter (QC) as a low weight and low-cost autonomous flight capable Unmanned
Aerial Vehicle (UAV) for delivering parcel ordered by online by using an android device
as its core on-board processing unit. This QC by following Google map can locate and
navigate destination. This paper demonstrates the QCs capability of delivering parcel
ordered by online and coming back to the starting place.

The promising result of this method enables future research on using QC for
delivering parcel. Introduction-As the demand for commercial deliveries increases within
cities, companies face a fundamental limitation in surface road capacity. Drone delivery
aims to overcome that limitation by exploiting the vertical dimension above city streets.
This report explores the vehicle design aspects of the delivery drone problem, including
flight efficiency, energy consumption, noise, and safety, which are central to the viability
of delivery drones. Importantly, key design constraints and expected performance levels
also speak to the potential scalability of the concept.

The model takes in a host of vehicle and mission parameters and assumptions, chief
among which are the aerodynamic, structural, and propulsive efficiencies and the battery
energy density.

DEPT. OF ISE, NCET, BENGALURU 2021-2022 8

Drone Delivery with Object Detection

CHAPTER-3
SYSTEM REQUIREMENTS AND SPECIFICATIONS

Every project has specific goals and serves particular purposes. Each goal and
purpose translate a process or several processes that the software/hardware aims to solve
or to automate. To deliver the right product, we should define well the software from the
beginning.
System requirement specification or SRS frameworks software development, it
documents every operation and dictates how software should behave, it can be as detailed
as what a button should doand should be as complete and correct as possible. The purpose
of a specification document is to describe the behavior as well as the different
functionalities of an application or software in a specific environment.

3.1 Hardware Requirements

1. Drone frame
2. Drone FC and ESC
3. FPV camera
4. Video module
5. Video antenna
6. Motors
7. Drone finder buzzer
8. Drone GPS module
9. Drone receiver
10. Remote
11. Module
12. Charger
13. Battery
14. esp32 cam module
15. Arduino uno R3
16. SIM900A GSM GPRS Module with RS232 Interface and SMA Antenna

DEPT. OF ISE, NCET, BENGALURU 2021-2022 9

Drone Delivery with Object Detection

3.2 Drone Components And Mechanism

A. Frame

The frame constitutes the skeleton of the drone and hosts all the other components.
When choosing a frame, particular attention should be devoted to its weight, size and
material, since these variables are related to the choice of other elements such as motors
and battery (e.g., for providing enough power for take-off), propellers (that should be able
to twist without interfering with each other), the number of components to be hosted on
the frame, etc. Frames are generally made of carbon fiber, thus being light though stiff.

B. Motors

The objective of motors is to spin the propellers, which are fastened on the top of
them. There are a lot of motors on the market, which vary, apart from their price, on the
basis of their spin speed at a constant voltage. The selection of a motor is linked to drone
weight, and influences the choice of other components such as the battery or the
Electronic Speed Controllers (ESCs). In the case of a quadcopter, four motors are needed.

C. Electronic Speed Controller

The Electronic Speed Controller is the component that communicates to the motors
how fast to spin in each situation, based on the signals received from the flight controller.
Each ESC controls a single motor; hence, for a quadcopter, four ESC should be used.
ESCs are then directly connected to the battery or to the power distribution board (which
acts as an intermediary between the battery and the different electronic components and,
for sake of simplicity has not been shown in the figure). In addition, many ESCs have a
built-in Battery Eliminator Circuit (BEC); in this way, they could provide power to other
components (mainly the flight controller and the radio receiver). The choice of the ESC
to purchase is critical, since the behaviour of each motor should be continuously adjusted.
For this reason, ESCs with a high refresh rate (more checks for new instructions from the
flight controller per second) should be preferred.

D. Flight Controller

The flight controller could be seen as the “brain” of the drone. In fact, it determines
the spin speed of each motor based on user’s commands and/or sensors’ data (flight

DEPT. OF ISE, NCET, BENGALURU 2021-2022 10

Drone Delivery with Object Detection

controllers usually contain several sensors such as gyroscopes for orientation,

accelerometers for acceleration, barometers for altitude, GPS for position, etc.).

E. Radio Receiver

The radio receiver and the transmitter (not reported in the figure) are the
components that allow the user to pilot the drone. When choosing them, particular
attention should be devoted to the number of channels. For multi-rotor control, they
should be at least four, being each control input (i.e., in the basic configuration, pitch,
roll, throttle and yaw) conveyed on a single channel.

F. Battery

Most quadcopters use LiPo batteries, that are lightweight and compact. The choice
of the battery should be made on the basis of drone motors and desired flight time. It is
worth considering that an increase of battery size is generally not proportional to the
increase of flight time, because of the greater weight of the battery.

G. Propellers

In a quadcopter, four propellers are needed. Propellers should be chosen by

considering the drone’s arm length, and the motors’ speed (e.g., for higher spin speeds
smaller propellers are suggested). Finally, it could be worth remarking that, in the case of
quadcopters, two propellers spin counter-clockwise and two spin clockwise; hence,
particular attention should be devoted in the assembly phase.

DEPT. OF ISE, NCET, BENGALURU 2021-2022 11

Drone Delivery with Object Detection

CHAPTER-4
SYSTEM ANALYSIS

4.1 Purpose of Research

The purpose of this paper is to propose a safe and secure delivery method while
providing a rapid delivery. This paper discusses a method where the camera inside the
cabin detects the object and notifies both sender and receiver with the details of the object.
Thus, ensures the safety of delivery and avoids fraud. This method can be adapted for
emergency delivery. This method might eventually become the primary delivery method.
Existing System:
The existing system is a drone that delivers products to the specified location
although it is not in existence in India. An autonomous drone is used. Security is still a
challenge.
Proposed System:
We propose a model that detects the object (product for delivery) and notify both
sender and receiver the details of the object with a text message. In this model, we aim to
build a manual drone which detects the objects and delivers faster which is easier for
distant delivery.

4.2 Advantages and Applications

Advantages:
• It is fast and reliable
• It has the features on in hand delivery
• It is robust and spontaneous
• It will avoid thefts and Hassel faced by customer
• Very efficient over short distances
• Less Expensive
• Safe and Secure delivery
Application:
It can be used any Field like:

DEPT. OF ISE, NCET, BENGALURU 2021-2022 12

Drone Delivery with Object Detection

• Military: To transfer the weapons or messages secretly

• Medical: To transfer any medicine Urgently
• Business: Delivering the Products securely & very fast

4.3 Obstacles To Overcome By Drone Delivery

For many experts however, there are still some major challenges, to be navigated before
drone technology become a basic part of business operations, such as:
1. Drones are still relatively light-weight, and can handle short flight durations of
15-30 minute flights with 10-15 pound payloads.
2. Drones also still represent an expensive investment for many companies, while
promises of savings are great in theory.
3. Dealing with the air traffic requirements of controlling flying objects.

DEPT. OF ISE, NCET, BENGALURU 2021-2022 13

Drone Delivery with Object Detection

CHAPTER-5
SYSTEM DESIGN

5.1 Methodology

The use of drones in last mile delivery introduces new issues to handle. The future
challenge underlying this application is not so much the design of drones for parcel
delivery, but to provide the security to customer (i.e., sender/receiver).
We need object detection module that can detect what is inside the drone and
notify user with its details with a text message thus eliminating the need for security.
We need esp32 cam module, Arduino uno R3, SIM900A GSM GPRS Module with
RS232 Interface and SMA Antenna, and drone frame and receiver for carrying out with
the project.
To detect the object, this model uses PIR infrared sensor which is placed in the
delivery cabinet. ESP32 cam module captures the image of the object with Arduino uno
r3, and PIR sensor connected with SIM900A GSM GPRS Module with RS232 Interface
to send the text message through telegram bot. The sensor detects when the object is
placed and notifies the end users with captured image of object and a message.

5.2 Design And Workflow

• Flowchart
As shown in figure 5.2.1, there are few simple steps in this method:
1. Detect the object
2. Send text message to sender and receiver
3. Take off from the base location
4. Go to delivery location
5. Drop the package in the location
6. Return to the base location

DEPT. OF ISE, NCET, BENGALURU 2021-2022 14

Drone Delivery with Object Detection

Figure 5.2.1: Flowchart of working model

• Working:
The base or the warehouse receives signal for the dispatch of the package. We use
an object detection module which detects the object/product that is kept in drone cabinet.
The details of this product are sent to the receiver through a text message to achieve safe
and secure delivery. Now the drone is all set to deliver the package to the destination.
The operator or pilot controls the drone.
It flies out rising to a higher altitude to avoid obstacles and disturbances. After it
flies to the higher altitude, it is lead to travel in the destined path. It will fly to the delivery
point and confirm the package to the correct customer and delivers. Since it is a fully
manual drone, the operator has total control throughout the process.
• Drone Base:
This is an abstract of a drone base created in a particular area that will operate in a
specific zone. In case the delivery location is out of its boundary it will communicate with
nearest bases. Then it delivers the package to the nearest base to the delivery location.
The base communicates with the customer and sends a drone with the required package
on its path to deliver it to the customer. The drone will deliver the package to the customer
only if it is 100 meters within the delivery location (the location provided at the time of
order).

DEPT. OF ISE, NCET, BENGALURU 2021-2022 15

Drone Delivery with Object Detection

• User System Interaction:

Long distance vehicles carrying more payloads require more energy and batteries
for each unit of travel. Larger multi-stop vehicles carry more battery and payload and pay
for it in drag. Vehicles are conservatively sized to be able to traverse the entire delivery
route and return to the fulfilment centre with all packages. The burden of irreversible
packages increases with each additional package and with compounds. On the delivery
date the information is sent to the customers phone and he updates the system with either
his live location. The customer receives the details of the package.

Figure 5.2.2: Flowchart of extracting user’s location

• Extraction Of User’s Location:
While ordering the customer will be asked to provide their address. They will have
the option to provide their live location or enter their location manually. In both cases

DEPT. OF ISE, NCET, BENGALURU 2021-2022 16

Drone Delivery with Object Detection

they have to set their location on the map provided. The drone will have unique design
and it will remove obstacles that use image processing to detect any obstacles and be safe
during storms and inclement weather. It will reach the people and once reached the
location it can be lowered by hand gestures.
This model uses a first person view drone (FPV) which is an UAV with a camera
which transmits the live video feed to mobile device, goggles, a headset or any another
display devices. The pilots of FPV drones see what the drone sees, by using this drone
we ensure the proper pickup and dropping of the package.

DEPT. OF ISE, NCET, BENGALURU 2021-2022 17

Drone Delivery with Object Detection

CHAPTER-6
IMPLEMENTATION

6.1 Model Components And Working

For this Project we’ve used FPV (First Person View) Drone with fully manual
operation. When flying an FPV drone, this basically means that the pilot sees what the
drone sees. Pilots operate "drones" with cameras while wearing head-mounted displays
that show the live stream camera feed from the drones in First Person View, or FPV. To
assemble a FPV drone we need several parts. Usually, FPV drones are either H or X
shaped. We assemble a X shape drone so that it’s easier to place a delivery cabinet for the
drone. A typical FPV drone parts are as mentioned below:

1. 4 Propellers
2. 4 Motors
3. Frame
4. Electronic Speed Controller (ESC)
5. Flight Controller
6. Lithium-Polymer Battery
7. Antenna
8. Radio Control Transmitter and Receiver

Figure 6.1:Components of Drone Model

All of these parts are assembled together to form a FPV drone. Frame is the
backbone of the multirotor, it is where all the components mount onto. Flight controller

DEPT. OF ISE, NCET, BENGALURU 2021-2022 18

Drone Delivery with Object Detection

(FC) is the brains of the multirotor, essentially the most complex component of the flight
system. ESC is the powerful component in the system, it commands the motor to spin,
direction of spin, and when to stop. Motors provide the power into your multirotor, it is
classified as brushless motors. Propellers are the blades that are connected to the motor
shaft, which spins just as the motors do. We have 4 propellers on each side of the
multirotor. The radio receiver receives the commands from pilot’s radio and pushes the
command to flight controller.

The drone is now a fully manual drone, pilot uses a remote to operate the drone, it
has a control range of 40 km i.e., LOS of front camera in the drone is 40km. The pilot can
see everything in a 40km range from where drone is. The flight time of the drone is 7-8
mins with a maximum speed of 140-180 kph. Drone uses a Li-Po battery which
rechargeable. The drone now weighs 650 gm.

Additionally, for delivery mechanism, we have included a delivery cabinet to the

model which has a payload of 500 gm and flight time of 5 mins with payload. To add a
cabinet, we’ve included a servo of 9 gm. The cabinet’s servo is used to close and open
the door of the cabinet.

6.2 Introduction to ESP32 CAM

The ESP32-CAM is a small size, low power consumption camera module based
on ESP32. It comes with an OV2640 camera and provides onboard TF card slot. The
ESP32-CAM can be widely used in intelligent IoT applications such as wireless video
monitoring, Wi-Fi image upload, QR identification.

Figure 6.2: ESP32-CAM Module

DEPT. OF ISE, NCET, BENGALURU 2021-2022 19

Drone Delivery with Object Detection

ESP32-CAM can be widely used in various IoT applications. It is suitable for home
smart devices, industrial wireless control, wireless monitoring, QR wireless
identification, wireless positioning system signals and other IoT applications.

6.3 Introduction to Arduino UNO

The Arduino UNO is the best board to get started with electronics and coding. If
this is your first experience tinkering with the platform, the UNO is the most robust board
you can start playing with. The UNO is the most used and documented board of the whole
Arduino family.

Figure 6.3: Arduino UNO board

Arduino UNO is a microcontroller board based on the ATmega328P. It has 14

digital input/output pins (of which 6 can be used as PWM outputs), 6 analog inputs, a
16 MHz ceramic resonator, a USB connection, a power jack, an ICSP header and a reset
button. It contains everything needed to support the microcontroller; simply connect it
to a computer with a USB cable or power it with an AC-to-DC adapter or battery to get
started.

6.4 Introduction to PIR Sensor

PIR sensors allow you to sense motion, almost always used to detect whether
a human has moved in or out of the sensors range. They are small, inexpensive, low-

DEPT. OF ISE, NCET, BENGALURU 2021-2022 20

Drone Delivery with Object Detection

power, easy to use and don't wear out. For that reason, they are commonly found in
appliances and gadgets used in homes or businesses. They are often referred to as
PIR, "Passive Infrared", "Pyroelectric", or "IR motion" sensors.
PIRs are basically made of a pyroelectric sensor (which you can see below as
the round metal can with a rectangular crystal in the centre), which can detect levels
of infrared radiation. Everything emits some low level radiation, and the hotter
something is, the more radiation is emitted. The sensor in a motion detector is
actually split in two halves. The reason for that is that we are looking to detect motion
(change) not average IR levels. The two halves are wired up so that they cancel each
other out. If one half sees more or less IR radiation than the other, the output will
swing high or low.

Figure 6.4: PIR Sensor

For many basic projects or products that need to detect when a person has left
or entered the area, or has approached, PIR sensors are great. They are low power
and low cost, pretty rugged, have a wide lens range, and are easy to interface with.

DEPT. OF ISE, NCET, BENGALURU 2021-2022 21

Drone Delivery with Object Detection

6.5 Modules

For better understanding of mechanism of the model, it can be explained in three

modules; namely, Object detection, Message Alert and Delivery module. Each module
uses specific technology and IOT devices for the working.

6.5.1 Object Detection Module

Figure 6.5.1: ESP32 CAM to Arduino UNO board

The ESP32-CAM is a small, compact ESP32 development board with a built-in

camera. It also has GPIO pins to connect it with few modules and a Micro SD card slot
for reading and saving images and data. We program both PIR sensor and ESP32 board
using Arduino IDE.

As shown in figure 6.5.1, we have used an ESP32-CAM to construct a motion

sensor detector with photo capture in this model. When your PIR sensor detects motion,
it awakens, takes a picture, and stores it in the micro SD card.

6.5.2 Message Alert Module:

Figure 6.5.2: Arduino UNO board to GSM SIM800L

DEPT. OF ISE, NCET, BENGALURU 2021-2022 22

Drone Delivery with Object Detection

We used SIM 800L module which is a miniature cellular module that allows for
GPRS transmission and to send and receive messages. The SIM 800L is interfaced with
Arduino UNO board. Once the object is detected, an SMS is sent to the receiver & the
images which are captured will be sent to a software application of the mobile. This
project makes use of GSM technology to communicate between mobile and embedded
devices.

6.5.3 Delivery Module:

Figure 6.5.3: Working procedure of delivery module

This module refers to moving product from point A to point B, and back to point
A. This model is manually controlled by the pilot. The model that we’ve built has a
payload of 500gm capacity, with 140 kph speed. The flight time is 15 mins with payload.
Once the product is received from the sender the pilot directs the drone to the delivery
location. Once at the receiver’s destination, the pilot slows down the drone, descends to
delivery approximately 5-7 meters from the ground and then releases the product in the
cabinet in the delivery point. Once the product is dropped in the destination, the drone
takes off from the destination and its way back to the pilot/sender.

The receiver needs ensure to collect the product when dropped. The only
requirement from the end user in this module is to ensure there is enough place to drop
the package.

DEPT. OF ISE, NCET, BENGALURU 2021-2022 23

Drone Delivery with Object Detection

CHAPTER-7
RESULT ANALYSIS AND DISCUSSION

Here in the implementation, we’ve built a drone. Drone is a complex device that
integrate different components working together and of course each component serve a
different function. Hence different materials are to be selected to build a certain part.There
are several important parts that can be break apart to explain the design better.

Such as the frame of the drone, the motors and propellars that produce the upward
thrust, the batteries which provide electrical energy to the drone, and sensors which play
a biggest role as the drone nervous system.

In our project, we manually flew a drone to deliver medications to the specified

location. A prototype system for a drone-based delivery service is discussed in this
research. Safe and secure delivery was considered which was achieved by sending
product details to customer. Below is the working model of drone.

Drone is a complex device that integrate different of components working together

and of course each component serves a different function. Hence different materials
are to be selected to build a certain part. There are several important parts that can be
break apart to explain the design better. Such as the frame of the drone, the motors and
propellers that produce the upward thrust, the batteries which provide electrical energy
to the drone, and sensors which play a biggest role as the drone nervous system.

DEPT. OF ISE, NCET, BENGALURU 2021-2022 24

Drone Delivery with Object Detection

Figure 7.1 Working model of drone

Figure 7.2 Drone Cabinet

DEPT. OF ISE, NCET, BENGALURU 2021-2022 25

Drone Delivery with Object Detection

Figure 7.3 Product for delivery

Figure 7.4 Connections on UNO Board

DEPT. OF ISE, NCET, BENGALURU 2021-2022 26

Drone Delivery with Object Detection

Figure 7.5 Bot Testing

DEPT. OF ISE, NCET, BENGALURU 2021-2022 27

Drone Delivery with Object Detection

CHAPTER-8
CONCLUSION AND FUTURE SCOPE

The aspect of this project is the future of technology. The manual delivery will not
be able to satisfy the demands of human race in the future. Drones will become a necessity
and will be of immense use in delivery of packages and couriers to the respective
customers. In future, people will become so engrossed in their own work that they would
not have time for collecting parcels and packages separately. Thus, these drones will
ensure the correct delivery of their demands in their current positions that is wherever
they are.
This method has successfully delivered a product for emergency purposes, but it
can also be utilized for other objectives, such as online shopping or military purposes,
particularly in a country like India, where logistics encounter numerous challenges on the
ground. Although the primary focus of our research was on ensuring secure delivery, this
module assures the model's safety by sending an SMS when the product is discovered,
preventing fraud and incorrect delivery. However, the future scope remains of designing
an autonomous drone.

DEPT. OF ISE, NCET, BENGALURU 2021-2022 28

Drone Delivery with Object Detection

REFERENCES

[1] Jaihyun Lee, Optimization of a Modular Drone Delivery System, IEEE, 1 May
2017, 2017 Annual IEEE International Systems Conference, 1-8.
[2] H. D. Yoo , S. M. Chankov , Drone-delivery Using Autonomous Mobility: An
Innovative Approach to Future Last-mile Delivery Problems, IEEE, 16-19
December 2018, 2018 IEEE International Conference on Industrial Engineering
and Engineering Management.
[3] Gino Brunner, Bence Szebedy, Simon Tanner and Roger Wattenhofer ,The Urban
Last Mile Problem: Autonomous Drone Delivery to Your Balcony, arxiv, 2019
International Conference on Unmanned Aircraft Systems (ICUAS) Atlanta, GA,
USA, June 11-14, 2019.
[4] Koki Yakushiji, Fumiatsu Yakushiji, Quality Control of Red Blood Cell Solutions
for Transfusion Transported via Drone Flight to a Remote Island, MDPI, Volume
5, Issue 3, 13 September 2021.
[5] Aditya Bhardwaj, Achira Basu, Package Delivery System Using GPS Drones,
International Journal of Engineering Research & Technology (IJERT) ISSN: 2278-
0181, Volume 08, Issue 10, 8 November 2019.
[6] Aniket Magdum ,Vivek Nikam-Patil ,Smart Drone Delivery System, International
Journal of Scientific Research in Engineering and Management (IJSREM),
Volume: 04 Issue: 03 Mar -2020.
[7] Johannessen, K.A.; Wear, N.K.S.; Toska, K.; Hansbø, M.; Berg, J.P. Fosse, E.
Pathologic Blood Samples Tolerate Exposure to Vibration and High Turbulence in
Simulated Drone Flights, but Plasma Samples Should be Centrifuged After Flight.
IEEE J. Transl. Eng. Health Med. 2021, 20, 4000110. [CrossRef]
[8] Sanjab, W. Saad, and T. Basar, “Prospect theory for enhanced cyberphysical
security of drone delivery systems: A network interdiction game,” in IEEE
International Conference on Communications, May 21-25, 2017, pp. 1–6.
[9] J. Stolaroff, C. Samaras, E. O’Neill, A. Lubers, A. Mitchell, and D. Ceperley,
“Energy use and life cycle greenhouse gas emissions of drones for commercial
package delivery,” Nature Communications, vol. 9, no. 1, 2018.

DEPT. OF ISE, NCET, BENGALURU 2021-2022 29

Drone Delivery with Object Detection

[10] M. M. Herrmann, “A Comparison of Unmanned Aerial Vehicle Regulations in the

United States and Europe,” , 53rd ASC Annual International Conference
Proceedings , pp. 299–306, 2017.
[11] C. C. Murray and A. G. Chu, “The flying sidekick traveling salesman problem:
Optimization of drone-assisted parcel delivery,” ScienceDirect, 2015, May 2015,
Pages 86-109.
[12] Garcia, I.Q, Velez, N.V, Alcaraz Martínez, P, Vidal Ull, J, Fernández Gallo B, A
quickly deployed and UAS-based logistics network for delivery of critical medical
goods during healthcare system stress periods: A real use case in Valencia (Spain),
MDPI, Volume 5, Issue 1, 17 February 2021, 5, 13.
[13] S. M. Adams, and C. J. Friedland. “A survey of unmanned aerial vehicle (UAV)
usage for imagery collection in disaster research and management.”,
Environmental Science, In Proc. of 9th Int. Workshop on Remote Sensing for
Disaster Response. 2011.
[14] D. Bamburry. "Drones: Designed for Product Delivery", Design Management
Review, vol. 26, no. 1, pp. 40-48, 2015, 23 July 2015.
[15] K. Dorling, J. Heinrichs, G. G. Messier, and S. Magierowski, “Vehicle Routing
Problem for Drone Delivery”, IEEE Transactions on System Man and Cybernetic:
Systems, 2016.
[16] N . Agatz, P. Bouman, and M. Schmidt, “Optimization Approaches for the
Traveling Salesman Problem with Drone”, ERIM Report Series Reference No.
ERS-2015-011-LIS, June 27, 2016.
[17] S. M. Ferrandez, T. Harbison, T. Weber, R. Sturges. “Optimization of a truck-
drone in tandem delivery network using k-means and genetic algorithm”, Journal
of Industrial Engineering and Management, vol. 9, no 2, pp. 374-388, 2016.
[18] A . E. Bayrak, A. X. Collopy, B. I. Epureanu and P. Y. Papalambros, "A
computational concept generation method for a modular vehicle fleet design",
2016 Annual IEEE Systems Conference (SysCon), Orlando, FL, pp. 1-8, 2016.
[19] R. Botsman, “Defining The Sharing Economy: What Is Collaborative
Consumption–And What Isn’t?,” 2015.
[20] F. Nex, and F. Remondino. "UAV for 3D mapping applications: a review" Applied
Geomatics, vol. 6, no. 1 pp. 1-15, 2014.

DEPT. OF ISE, NCET, BENGALURU 2021-2022 30

Drone Delivery with Object Detection

[21] V. C. Usenko, L. von Stumberg, A. Pangercic, and D. Cremers, “Real-time

trajectory replanning for MAVs using uniform bsplines and a 3d circular buffer,”
in IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS,
Vancouver, BC, Canada, September 24-28, 2017, pp. 215–222.
[22] UPS, “UPS tests residential delivery via drone launched from atop package car”,
2018, accessed: 2019-05-09.

DEPT. OF ISE, NCET, BENGALURU 2021-2022 31

Mr. H. Hemanth Kumar
Executive Secretary
11th May, 2022
Ref: 7.1.01/SPP/91

The Principal,
Nagarjuna College of Engineering & Technology,
Mudugurki, Venkatagiri Kote,
Devanahalli,
Bengaluru - 562 110

Dear Sir/Madam,

Sub : Sanction of Student Project - 45th Series: Year 2021-2022

Project Proposal Reference No. : 45S_BE_1768

Ref : Project Proposal entitled DRONE DELIVERY WITH OBJECT DETECTION

We are pleased to inform that your student project proposal referred above, has been approved by
the Council under "Student Project Programme - 45th Series". The project details are as below:
Student(s) Ms. MOGALIKUDHURU SUMA RANI
INFORMATION SCIENCE AND
Ms. MATTAPARTHI LIKHITHA SHREE Department
ENGINEERING
Mr. NIROOP S RAO
Ms. ANANYA VIKRAM Sanctioned
Guide(s) Mrs. SANDHYA L Amount 6,000.00
(in Rs.)

Comments / Suggestions of the Experts

Build suitable algorithm for auto detection system

Instructions:
a) The project should be performed based on the objectives of the proposal submitted.
b) Any changes in the project title,objectives or students team is liable for rejection of the project
and your institution shall return the sanctioned funds to KSCST.
c) Please quote your project reference number printed above in all your future correspondences.
d) After completing the project, 2 to 3 page write-up (synopsis) needs to be uploaded on to the
following Google Forms link https://forms.gle/YMn9K7XETu96i8KbA. The synopsis should
include following:
1) Project Reference Number
2) Title of the project
3) Name of the College & Department
45S_BE_1768
4) Name of the students & Guide(s)
5) Keywords
6) Introduction / background (with specific reference to the project, work done earlier, etc) -
about 20 lines
7) Objectives (about 10 lines)
8) Methodology ( about 20 lines on materials, methods, details of work carried out, including
drawings, diagrams etc)
9) Results and Conclusions (about 20 lines with specific reference to work carried out)
10) Scope for future work (about 20 lines).
e) In case of incompeted projects, the sanctioned amount shall be returned to KSCST.
f) The sanctioned amount will be transferred by NEFT to the bank account provided by the
College/Institute.
g) The sponsored projects evaluation will be held in the Nodal Centre/Online Mode and the details
of the same will be intimated shortly by email / Website announcement.
h) After completion of the project, soft copy of the project report duly signed by the Principal, the
HoD, Guide(s) and studetn(s) shall be uploaded in the following Google Forms Link
https://forms.gle/PciAaAVisn6bn8AM7. The report should be prepared in the format prescribed
by the university.

Please visit our website for further announcements / information and for any clarifications please
email to [email protected]

Thanking you and with best regards,

Yours sincerely,

(H. Hemanth Kumar)

Copy to:
1) The HoD
INFORMATION SCIENCE AND ENGINEERING
NAGARJUNA COLLEGE OF ENGINEERING AND TECHNOLOGY, BENGALURU

2) Mrs. SANDHYA L
INFORMATION SCIENCE AND ENGINEERING
NAGARJUNA COLLEGE OF ENGINEERING AND TECHNOLOGY, BENGALURU

3) THE ACCOUNTS OFFICER

KSCST, BENGALURU
GRADIVA REVIEW JOURNAL ISSN NO : 0363-8057

Emergency Medicine Delivery System Using Drone with

IoT

Sandhya L1, Muthuraju V2, Niroop S Rao3, Mogalikudhuru Suma Rani4 and
Mattaparthi Likhitha Shree5
1,2
Faculty, Department of Information Science and Engineering, Nagarjuna
College of Engineering and Technology, Bengaluru, Karnataka, India
3,4,5
Student, Department of Information Science and Engineering, Nagarjuna
College of Engineering and Technology, Bengaluru, Karnataka, India

Abstract: Drones, or unmanned aerial vehicles (UAVs), have been used successfully in
disaster relief, 3D mapping, and precision agriculture. A delivery drone is an unmanned
aerial vehicle that transports deliveries, medications, food, and other items. Typically,
delivery drones are autonomous. Recently, drones are considered as an effective key to
express delivery. Drones will be delivering things in the near future, however there are
concerns about receiving damaged parcels, interference from people and weather, and
packages being delivered to the wrong address. This project is mainly concerned with
safe and secure delivery. We have created a drone that can transport important and vital
medications in times of emergency. The main objective of our model is to detect the
medicine kept inside the drone cabin and notify receiver the details of the object with the
use of esp32 cam and sim 800l module. This drone can be very helpful in emergency
situations and in places where there are still gaps in the availability of reliable
transportation services.

Keywords: Unmanned Aerial Vehicle, Drones, delivery, medicine, emergency, message, UAV,
FPV, Esp32

1. Introduction

The evolution in technology over the last century has changed people’s approach in
performing daily activities. This is often evident when considering, washing machines,
dish washers, vacuum cleaners, etc. We observe the focus on "drones" (unmanned
aircrafts directed by remote control) and Unmanned Aerial Vehicles (aircrafts piloted by
remote control) which are now widely available in the commercial market, among many
other advancements. Drones and unmanned aerial vehicles (UAVs) may soon
revolutionize delivery services. A delivery drone typical design is about what is being
delivered, and where it must be delivered. As a result, the use of drones in last-mile
delivery raises additional concerns that must be addressed. Not so much the design of
drones for parcel delivery rather the future problem underpinning this application, but to
provide the security to customers (i.e., sender/receiver). Also, there have been numerous
cases documented in India where the delayed delivery of medications to any healthcare
facility proved to be extremely lethal. This drone delivery project will be helpful in cases
of efficient medicine delivery in cities where traffic congestion and poor road conditions
become a major drawback in all the aforementioned situations.

VOLUME 8 ISSUE 7 2022 PAGE NO: 178

GRADIVA REVIEW JOURNAL ISSN NO : 0363-8057

2. Related Work

The paper[1] introduces modular architecture to a drone delivery system. A fully modular
drone system is compared to a non-modular drone system in this article. One of the
reasons to introduce modular architecture is the limited battery capacity in drones. The
proposed solution is compared to both static and dynamic management strategies. All of
the strategies are tested on modular as well as non-modular drones, and the results
demonstrate that the static management strategy makes the drone delivery system unstable
for both systems. Although both dynamic programming and the proposed approach
proven to be stable for drone delivery, the proposed method saves time and energy. In
comparison to a non-modular system, the results show that a modular drone delivery
system has the potential to improve delivery time. However semi modular system with
replaceable batter was not analyzed.

The author of paper[2] suggests a novel delivery idea known as Drone-Delivery with
Autonomous Mobility (DDAM). DDAM was used to improve delivery efficiency and
speed up the process. The major goal of this approach is to deliver packages from the
Distribution center to the Delivery location. Drones carry packages using autonomous
mobility vehicles as an intermediary mode of transportation. This concept has the
potential to improve delivery speed and flexibility. Concept development has its
constraints. This proposal was considered to be more practical as an alternative delivery
technique rather than the primary delivery method after multiple assessments and
suggestions. DDAM, on the other hand, is still a concept with futuristic features that has
yet to be marketed.

This concept was not compared to any other concepts statistically. However, limitations
of this concept are, drones which connect to autonomous mobile vehicle uses road way as
the path, which takes the same time as any delivery agent. There’s a need of GPS location
on both the sides (Distribution center to Delivery destination), till the delivery is
completed.

The paper[3] focuses on autonomous last mile delivery in urban environments. The author
aims to provide the delivery to the balcony of the house/apartment of the recipient. This
system was developed based on open-source project Robot Operating System (ROS). The
ROS package allows drone to use its front facing camera. The drone looks for the mark on
door/wall of the balcony of the delivery location. This prototype uses GPS to fly to an
approximate delivery location and enables visual navigation to identify the exact delivery
position (Visual Marker Tracker). The main purpose of this paper is to gain collision
avoidance, while this drone uses front facing camera which can be complicated while
navigating in tight spaces and landing. This drone is working on visual data for navigating
to destination. However, the main limitation of this paper is, it uses an autonomous drone
with a front facing camera, which creates complications while navigating and landing.

The goal of the study[4] was to use drones to distribute blood packs. The authors used a
drone to ferry red blood cells (RBCs) packets from Sasebo to Arikawa port over the sea.
Temperature was the most crucial element that could alter RBC quality during shipment.
For a transfusion sent to a remote island, quality control of the RBC solution was done
utilizing a drone. The cross-matching tests yielded no surprises. The fact that the weather

VOLUME 8 ISSUE 7 2022 PAGE NO: 179

GRADIVA REVIEW JOURNAL ISSN NO : 0363-8057

had no effect on the experiment while it was being carried was one of the reasons for its
success. This experiment shows that blood might be transported at controlled
temperatures using the UH transport. If blood was not utilized upon arrival, it can be used
before the expiration date if it was delivered at an adequate temperature. This is the first
time to transport transfusion packets of RBC solution beyond a part of the western Pacific
Ocean. However, safety and security were not taken into account.

The paper[5] uses GPS drone delivery system. It would detect the consumers location
through GPS and deliver the package as per the time. This drone includes a GPS and live
tracking system that can detect the user's location. To prevent fraud, an OTP technique is
used during package generation. This system acts like any home delivery app where the
location can be tracked on both sides to provide accurate delivery. This system is effective
in terms of user and system interaction. However, image processing can be done to
remove obstacles in the delivery path.

This paper[6] research examines a systematic online delivery procedure with an

independent quality control system. As its central processing unit, it uses an Android
device. QC would transport the product to the consumer using Google Maps, which in
turn saves time and reduce delivery manpower. The autonomous QC will pick up the
products at the pickup site, fly on, and deliver them to the respective location using the
co-ordinates provided. There is an extra aspect of this Autonomous QC i.e., it contains an
ultrasonic sensor that would avoid collisions by detecting the object before.

3. Purpose Of Research

The purpose of this paper is to propose a safe and secure delivery method while providing
a rapid delivery. This paper discusses a method where the camera inside the cabin detects
the object and notifies both sender and receiver with the details of the object. Thus,
ensures the safety of delivery and avoids fraud. This method can be adapted for
emergency delivery. This method might eventually become the primary delivery method.

Abbreviations And Acronyms

List of acronyms and abbreviations

1. UAV Unmanned Aerial Vehicle

2. QC Quadcopter
3. ROS Robot Operating System
4. FPV First Person View
5. GSM Global System for Mobile Communication
6. ESC Electronic Speed Controller
7. PIR Passive Infrared Sensor
8. GPRS General Packet radio service
9. FC Flight Controller
10. ESC Electronic Speed Controller
11. LOS Line Of Sight

4. Methodology
As shown in figure 4.1, there are few simple steps in this method:

VOLUME 8 ISSUE 7 2022 PAGE NO: 180

GRADIVA REVIEW JOURNAL ISSN NO : 0363-8057

1. Detect the object

2. Send text message to sender and receiver
3. Take off from the base location
4. Go to delivery location
5. Drop the package in the location
6. Return to the base location

This model uses a first person view drone (FPV) which is an UAV with a camera which
transmits the live video feed to mobile device, goggles, a headset or any another display
devices. The pilots of FPV drones see what the drone sees, by using this drone we ensure
the proper pickup and dropping of the package.

Figure.4.1: Working model of drone

To detect the object, this model uses PIR infrared sensor which is placed in the delivery
cabinet. ESP32 cam module captures the image of the object with Arduino uno r3, and
PIR sensor connected with SIM900A GSM GPRS Module with RS232 Interface to send
the text message through telegram bot. The sensor detects when the object is placed and
notifies the end users with captured image of object and a message.

5. Implementation
For this Project we’ve used FPV (First Person View) Drone with fully manual operation.
When flying an FPV drone, this basically means that the pilot sees what the drone sees.
Pilots operate "drones" with cameras while wearing head-mounted displays that show the
live stream camera feed from the drones in First Person View, or FPV. To assemble a
FPV drone we need several parts. Usually, FPV drones are either H or X shaped. We

VOLUME 8 ISSUE 7 2022 PAGE NO: 181

GRADIVA REVIEW JOURNAL ISSN NO : 0363-8057

assemble a X shape drone so that it’s easier to place a delivery cabinet for the drone. A
typical FPV drone parts are as mentioned below:

1. 4 Propellers
2. 4 Motors
3. Frame
4. Electronic Speed Controller(ESC)
5. Flight Controller
6. Lithium-Polymer Battery
7. Antenna
8. Radio Control Transmitter and Receiver

All of these parts are assembled together to form a FPV drone. Frame is the backbone of
the multirotor, it is where all the components mount onto. Flight controller(FC) is the
brains of the multirotor, essentially the most complex component of the flight system.
ESC is the powerful component in the system, it commands the motor to spin, direction of
spin, and when to stop. Motors provide the power into your multirotor, it is classified as
brushless motors. Propellers are the blades that are connected to the motor shaft, which
spins just as the motors do. We have 4 propellers on each side of the multirotor. The radio
receiver receives the commands from pilot’s radio and pushes the command to flight
controller.

The drone is now a fully manual drone, pilot uses a remote to operate the drone, it has a
control range of 40 km i.e., LOS of front camera in the drone is 40km. The pilot can see
everything in a 40km range from where drone is. The flight time of the drone is 7-8 mins
with a maximum speed of 140-180 kph. Drone uses a Li-Po battery which rechargeable.
The drone now weighs 650 gm.

Additionally, for drone delivery we have included a delivery cabinet to the model which
has a payload of 500 gm and flight time of 5 mins with payload. To add a cabinet, we’ve
included a servo of 9 gm. The cabinet’s servo is used to close and open the door of the
cabinet.

Object Detection Module:

Figure.5.1: ESP32 CAM to Arduino UNO board

VOLUME 8 ISSUE 7 2022 PAGE NO: 182

GRADIVA REVIEW JOURNAL ISSN NO : 0363-8057

The ESP32-CAM is a small, compact ESP32 development board with a built-in camera. It
also has GPIO pins to connect it with few modules and a Micro SD card slot for reading
and saving images and data. We program both PIR sensor and ESP32 board using
Arduino IDE.

As shown in figure.5.2., we have used an ESP32-CAM to construct a motion sensor

detector with photo capture in this model. When your PIR sensor detects motion, it
awakens, takes a picture, and stores it in the micro SD card.

SMS Alert Module:

Figure.5.2: Arduino UNO board to GSM SIM800L

We used SIM 800L module which is a miniature cellular module that allows for GPRS
transmission and to send and receive messages. The SIM 800L is interfaced with Arduino
UNO board. Once the object is detected, an SMS is sent to the receiver & the images
which are captured will be sent to a software application of the mobile. This project
makes use of GSM technology to communicate between mobile and embedded devices.

Delivery Module:

Figure.5.3: Working procedure of delivery module

This module refers to moving product from point A to point B, and back to point A. This
model is manually controlled by the pilot. The model that we’ve built has a payload of
500gm capacity, with 140 kph speed. The flight time is 15 mins with payload. Once the
product is received from the sender the pilot directs the drone to the delivery location.
Once at the receiver’s destination, the pilot slows down the drone, descends to delivery
approximately 5-7 meters from the ground and then releases the product in the cabinet in
the delivery point. Once the product is dropped in the destination, the drone takes off from
the destination and its way back to the pilot/sender.

VOLUME 8 ISSUE 7 2022 PAGE NO: 183

GRADIVA REVIEW JOURNAL ISSN NO : 0363-8057

The receiver needs ensure to collect the product when dropped. The only requirement
from the end user in this module is to ensure there is enough place to drop the package.

6. Results

In our project, we manually flew a drone to deliver medications to the specified location.
A prototype system for a drone-based delivery service is discussed in this research. Safe
and secure delivery was considered which was achieved by sending medicine details to
customer. Below is the working model of drone.

Figure.6.1: Images of working model

7. Conclusion & Future Scope

This method has successfully delivered a product for emergency purposes, but it can also
be utilized for other objectives, such as online shopping or military purposes, particularly
in a country like India, where logistics encounter numerous challenges on the ground.
Although the primary focus of our research was on ensuring secure delivery, this module
assures the model's safety by sending an SMS when the product is discovered, preventing
fraud and incorrect delivery. However, the future scope remains of designing an
autonomous drone.

References

[1] Optimization of a Modular Drone Delivery System Jaihyun Lee Department of

Mechanical Engineering University of Michigan Ann Arbor, MI, United States.
[2] Drone-delivery Using Autonomous Mobility: An Innovative Approach to Future Last-
mile Delivery Problems H. D. Yoo1, S. M. Chankov2 1 Department of Mathematics &
Logistics, Jacobs University Bremen, Bremen, Germany.
[3] The Urban Last Mile Problem: Autonomous Drone Delivery to Your Balcony, Gino
Brunner, Bence Szebedy, Simon Tanner and Roger Wattenhofer* Computer Engineering
and Networks Laboratory ETH Zurich, 2019 International Conference on Unmanned
Aircraft Systems (ICUAS) Atlanta, GA, USA, June 11-14, 2019.
[4] Quality Control of Red Blood Cell Solutions for Transfusion Transported via Drone
Flight to a Remote Island, Koki Yakushiji, Fumiatsu Yakushiji, https://doi.org/10.3390/
drones5030096.

VOLUME 8 ISSUE 7 2022 PAGE NO: 184

GRADIVA REVIEW JOURNAL ISSN NO : 0363-8057

[5] Package Delivery System Using GPS Drones, International Journal of Engineering
Research & Technology (IJERT) ISSN: 2278-0181, http://www.ijert.org .
[6] Smart Drone Delivery System, Aniket Magdum ,Vivek Nikam-Patil, International Journal
of Scientific Research in Engineering and Management (IJSREM), Volume: 04 Issue: 03
Mar -2020.

VOLUME 8 ISSUE 7 2022 PAGE NO: 185