Project Synopsis

on

“MEMORY CARD GAME ”

Submitted in complete fulfillment of the requirement for the degree of
Bachelors of Engineering by

Yahraj Tripathi
Ankur Singh
Anjali Samant
Sejal Ubale

Under the guidance of

Prof. Sapana Bhuskute

LOKMANYA TILAK COLLEGE OF ENGINEERING

Affiliated to
UNIVERSITY OF MUMBAI

Computer Science and Enginnering(Data Science)

Academic Year – 2024-2025
 CERTIFICATE
This is to certify that Yashraj Tripathi, Ankur Singh, Anjali Samant, Sejal
Ubale, have delivered seminar for Project Stage- II on 16th October 2024 and
submitted a report in Lokmanya Tilak College of Engineering, Navi Mumbai for the
partial fulfilment of the degree of B.E in “ Computer Science and Enginnering(Data
Science) from University of Mumbai, for the year 2024-25.

Prof. Sapana Bhuskute Dr. Nandini C Nag Dr. Subhash Shinde

(Project Guide) (Head of Department) (Principal)

External Examiner

Place: Lokmanya Tilak College of Engineering

Date: 16-10-2024

i
 ACKNOWLEDGEMENT

We would like to acknowledge and extend our heartfelt gratitude to all those people
who have been associated with this project and have helped us with it thus making
it a worthwhile experience.
Firstly we extend our thanks to various people which include our project guide
Prof. Sapana Bhuskute who has shared her opinions and experiences through
which we received the required information crucial for our project synopsis. We
are also thankful to head of the department Dr. Nandini C Nag and all the
staff members of Electronics & Telecommunication Department for their highly
co-operative and encouraging attitudes, which have always boosted us.
We also take this opportunity with great pleasure to thank our Principal Dr.
Subhash Shinde whose timely support and encouragement has helped us succeed
in our venture.

Name of Candidate Signature

1.Sanchit Kachale
2.Jitesh Kanojia
3.Kiran Patil
4.Deep Kirdat

ii
 ABSTRACT

The Memory Game is a concentration-based puzzle implemented using Java and

the Swing framework. Players are presented with a grid of hidden cards and must
match pairs by flipping two cards at a time. The game enhances memory and
cognitive skills, challenging players to remember the position of cards they have
previously revealed. Developed using object-oriented programming principles, the
game features a graphical user interface, a timer, and a scoring system that rewards
accuracy and speed. It can be extended with various difficulty levels, themes,
and multiplayer modes, showcasing Java’s capabilities in creating interactive and
engaging applications. .

iii
Contents

Acknowledgement ii

Abstract iii

List of Tables vi

List of Figures vii

List of Abbreviations viii

List of Symbols ix

1 INTRODUCTION 1
1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2.1 problem 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2 LITERATURE SURVEY 4
2.1 Referred Paper 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.2 Referred Paper 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.3 Referred Paper 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.4 Existing System & its Drawbacks . . . . . . . . . . . . . . . . . . . 6

3 PROPOSED METHODOLOGY 8
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

4 IMPLEMENTATION 12
4.1 Working Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4.2 Software and Hardware Used . . . . . . . . . . . . . . . . . . . . . . 14
4.2.1 Hardware . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

iv
 4.2.2 Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

5 RESULTS AND DISCUSSION 26

5.1 Top View of Bot . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

6 CONCLUSIONS 28
6.1 Project Highlights . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

7 ACHIEVEMENTS 29

8 EXTRAS 33

References 34

v
List of Tables

8.1 Pitch and Formants of Original and Reconstructed vowel /a/ . . . . 35

vi
List of Figures

3.1 Control Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

3.2 Transistor as a switch . . . . . . . . . . . . . . . . . . . . . . . . . . 11

4.1 System Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

4.2 Robot Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4.3 Node MCU Pinout . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
4.4 IR Wheel encoder sensor . . . . . . . . . . . . . . . . . . . . . . . . 16
4.5 L293D Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . 18
4.6 BO Motor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.7 Battery Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.8 Mini Submersible pump motor . . . . . . . . . . . . . . . . . . . . . 21
4.9 TIP41 Pinout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.10 TIP41 Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.11 LM7805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.12 Internal Block Diagram LM7805 . . . . . . . . . . . . . . . . . . . . 24

5.1 Top View of Bot . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

7.1 AVISHKAR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

8.1 NMF flowchart with Mixed signal matrix - M, Basis function matrix
- B, Weight matrix - W . . . . . . . . . . . . . . . . . . . . . . . . 34

vii
List of Abbreviations

ASNA . . . . Active-set Newton Algorithm

ASS . . . . . . Audio Source Separation
BSS . . . . . . Blind Source Separation
BSS EVAL . Blind Source Separation Toolbox
CASA . . . . Computational Auditory Scene Analysis
WER . . . . . Word Error Rate
WRR . . . . . Word Recognition Rate

viii
List of Symbols

t . . . . . . . . Number of time frames per basis vectors

M . . . . . . . Mixed Signal
κs . . . . . . . Scattering coefficient
δp . . . . . . . Penetration depth
τ . . . . . . . . Pulse length of SAR signal
β . . . . . . . . Bandwidth of the SAR signal
θr . . . . . . . Local refractive angle
δA . . . . . . . Azimuth resolution of the SAR data
s . . . . . . . . RMS surface roughness

ix
Chapter 1

INTRODUCTION

In mobile robotics, autonomous navigation is an important feature because it

allows the robot to independently move from a point to another without a tele-
operator. Numerous techniques and algorithms have been developed for this pur-
pose, each having their own merits and shortcomings. Maze-solving – although
artificial in terms of the confine that the robot it subjected to – is a structured
technique and controlled implementation of autonomous navigation which is some-
times preferable in studying specific aspect of the problem. This paper discusses
an implementation of a small size mobile robot designed to solve a maze based on
the flood-fill algorithm. The maze-solving task is similar to the ones in the Mi-
croMouse competition where robots compete on solving a maze in the least time
possible and using the most efficient way. A robot must navigate from a corner of
a maze to the center as quickly as possible. It knows where the starting location is
and where the target location is, but it does not have any information about the
obstacles between the two. The maze is normally composed of 256 square cells,
where the size each cell is about 18 cm × 18cm. The cells are arranged to form a
16 row × 16 column maze. The starting location of the maze is on one of the cells
at its corners, and the target location is formed by four cells at the center of the
maze. Only one cell is opened for entrance. The requirements of maze walls and
support platform are provided in the IEEE standard. There are many aspects to
the future of this Agribot. Agriculture is considered one of the most important

1
economic activities in India. The bot uses various techniques that help us track the
various activities involved in the farming process such as soil moisture level, soil
type, different nutrient levels in the soil, suggestion of the crop to be cultivated.
The multi functionality of the robot will also help the farmer use the same robot
to extract weeds, maintain records on soil data, and make it available at any time
as it will be stored in a cloud server. Farmers using bots will be easier to monitor
the field

1.1 Motivation
This project mainly focuses on the agricultural aspects whether it be beyond reach
farming or eliminating the adverse side effects of fertilizers and pesticides on farm-
ers health. Fertilizers and pesticides contain harmful chemical compostions which
can cause various health hazards not only limited to skin cancer, inflammation,
itching and permanent skin damage. The primary objective is to eliminate the
manual work along with safeguarding farmers from exposure of toxic chemicals
hence benifiting both health and time aspects. This compact bot can accompish
farming in difficult geographical terrains where human navigation is difficult.

1.2 Problem Statement

This project mainly focuses on the agricultural aspects whether it be beyond reach
farming or eliminating the adverse side effects of fertilizers and pesticides on farm-
ers health. Fertilizers and pesticides contain harmful chemical compostions which
can cause various health hazards not only limited to skin cancer, inflammation,
itching and permanent skin damage. The primary objective is to eliminate the
manual work along with safeguarding farmers from exposure of toxic chemicals
hence benifiting both health and time aspects. This compact bot can accompish
farming in difficult geographical terrains where human navigation is difficult.

2
1.2.1 problem 1
x 2 a2
(1.1)
y b41

3
Chapter 2

LITERATURE SURVEY

2.1 Referred Paper 1

A quick development of innovation moves us to plan the best choice for an accurate
mission. Numerous independent automated innovations are intimated in the lives
of individuals making their work much easier. It has been seen that automated
vehicles are presented so far, with shrewd abilities after enormous measures of cash
spent yearly on the examination. Here in this paper, autonomous maze solving
robot is developed with independent mapping and localization skill. Firstly, the
maze solving vehicle is designed with three infrared sensors of which two is used for
wall detection to avoid collision and the third is for obstacle detection for picking
and placing the objects to clear its pathway with the help of robotic arm. Also,
it desires to use robot where an environment unreachable for human. In addition,
there are also places where use of robots is the only way to achieve a goal. For
this, appropriate placement of sensory devices is very critical. We have successfully
implemented a maze solving ability onto the robot so called MazeBot. It has been
tested that the robot can solve the maze successfully without any interruption
with the walls and the objects. In this design, the accuracy of measurements
and the real-time processing allied with minimum processing power are the key
components in overall embedded design.

4
2.2 Referred Paper 2
The problem of micro-mouse is 30 years old but its importance in the field of
robotics is unparalleled, as it requires a complete analysis and proper planning to
be solved. This paper covers one of the most important areas of robot, “Decision
making Algorithm” or in lay-man’s language, “Robot Intelligence”. For starting
in the field of micro-mouse it is very difficult to begin with highly sophisticated
algorithms. This paper begins with very basic wall follower logic to solve the
maze. And gradually improves the algorithm to accurately solve the maze in
shortest time with some more intelligence. The Algorithm is developed up to
some sophisticated level as Flood-Fill algorithm. The paper would help all the
beginners in this fascinating field, as they proceed towards development of the
“brain of the system”, particularly for robots concerned with path planning and
navigation.

5
2.3 Referred Paper 3
This paper presents the design and development of a robot capable to navigate
through a simple maze, based on algorithms for maintaining the trajectory, making
turns and finding the exit from the maze. Once we have been able to create such
a robot, expanding the robot by adding new functionalities leads to its transfor-
mation, for example into a room vacuum cleaner, or why not, if the environment
is a labyrinth, in a tool capable of determining the shortest route from any two
points of the labyrinth. Its dimensions allow exploration of spaces inaccessible to
humans. Robots of this kind are based on decision making algorithms, of which
we mention the wall tracking algorithm and the flood fill algorithm. The robot
presented is based on the wall tracking algorithm running on a microcontroller.
This robot is made up of a printed circuit board that also has the role of chassis,
two DC gear motors with incremental encoders, six distance sensors and a battery.

2.4 Existing System & its Drawbacks

Problem with current scenario

• Inappropriate Knowledge: Contact with harmful fertilizers and pesticides

can have adverse effects on farmers skin and can cause breathing problems.

• Most of the readily available agricultural robots use heavy onboard compu-
tation which requires high battery capacity just for computation and which
is not being used for actual field work.

• Positioning and detecting number of turns is based on external factors in

existing topology such as line following robots this dependency on external
factors can be reduced.

• Control of the robot requires newer networks and specialized equipment

which furthermore increases cost of the whole system.

• The robots can only be controlled by manual means and lack interfacing with
automation hence reduced friction between automation and manual control
is created by using a single web based interface.

6
 • Use of the total power of the control board is very less hence a power hungry
board is not used to its maximum performance decreasing the bots efficiency.

• Less number of modes for a modular type of design limits the use of the
robot for multiple types of work.

• Lack of manpower: The wages of farm workers is less than construction

labors so for living they choose different path but not farming.

Therefore, there is a need to develop and implement a Agricultural robot which

deals with these drawbacks and provides a complete solution.

7
Chapter 3

PROPOSED METHODOLOGY

3.1 Introduction
India has always been an agricultural country and mainly depends on natural
resources. In the early days, farming was done manually by bullock cart which
was very difficult and time taking for the farmer. India being an agrarian coun-
try needs new agricultural techniques, this paperwork will definitely help in the
economic development of the country. Agribot will be the game changer in the
agricultural sector. The objective is to build a robot that has different farming
methods that can be adjusted according to requirement. Targeting in a key sector
in the Indian economy is harder than expected. The bot will have an adjustable
ground clearance mechanism which will make it a multi-purpose bot. Imagining a
robot with interchangeable arms and adjustable ground clearance gives us a huge
advantage over the counter parts available in the market. Technology in farming
can add more benefits for the farmer and the country. A bot in the field can do all
the tasks that cause adverse effects on farmers such as by spraying liquid fertiliz-
ers and pesticides. This will also save the time and effort of the farmer. Advance
robots can also monitor the farm with the help of sensors and keep farmers up-
dated on the current status of the farm. The future of farming will depend mostly
on agricultural bots.

8
3.2 Methodology
The Motors and chassis are decided according to the requirement of farmer and
field accordingly a motor driver is selected as well to deliver sufficient amount
of power required by the motors, if motors are high current then lower gauge
wires should be used as to keep any losses in the wires low. Also power hungry
motors require appropriate battery packs that decide the total runtime of the bot.
For maximum power density use of Lithium ion or Lithium Polymer batteries are
best suited although due to requirement of complex Battery Management System
(BMS) as the battery pack gets bigger use of a simple Lead Acid battery can
be used with simpler charging circuits. The Nodemcu uses 3.3V power which
is created by an Low Dropout Voltage Regulator (AMS1117-3.3) in a SOT-223
package which cant dissipate heat if the input voltage is greater than 10V to the
Vin pin of NodeMCU. As there may be high power dissipated by the voltage
regulator hence an external voltage regulator with higher power dissipation that
has a proper heatsink can be used for better thermal management of the board.
We use LM7805 (5V regulator) which uses a TO-220 package that is connected to a
heatsink. a 5V regulator is powering the 3.3V regulator on the NodeMCU control
board instead of directly using a 3.3V regulator to getter better noise immunity
of the Higher Voltage power line that generates Noise due to the High current
Inductive motors and other Inductive load switching such as the pump or relays.
The motors are powered by the higher voltage power rail driven by a motor driver
IC such as L293D which requires 4 input logical signals which is provided by the
NodeMCU to control the direction of each motor. The motors are not all identical
hence supplying equal amounts of electrical power to them will not result in the
same mechanical output power hence the use of time based control of motors for
turning or moving the robot forward will be dependent on not only time as well as
the battery power available. This problem is solved by using a IR wheel encoder
sensor attached to the geared motors to get the number of turns the wheel has
turned. This whole process is a closed control loop that makes sure the robot turns
precisely 90 degrees or moves certain number of steps forward or backwards.

9
 Figure 3.1: Control Algorithm

For programming the NodeMCU board the USB micro connector onboard is used
with the CP2102 IC for USB to TTL conversion. By using Arduino IDE and using
various libraries it provides for the ESP8266 we can easily implement a http web
server connected to a Local Network. The API’s are each individually hardcoded
and each have their own tasks that they do. For getting the number of turns
the wheel has turned instantly the digital output of the sensor is used to trigger
an hardware interrupt that just increments the number of wheel turns counter in
memory. A DNS server is also setup so that the bot can get commands directly
through a URL instead of requiring to know the local network IP.

The two motor pumps (Left Pump, Right Pump) can be controlled individually or
both together by using their respective API call (sprayleft, sprayright, sprayboth).
Each pump uses the higher voltage rail for power and is switched using a TIP41
NPN Darlington pair as a low side switch which is activated by the NodeMCU IO
pin connected to the transistors Base.

10
Figure 3.2: Transistor as a switch

11
Chapter 4

IMPLEMENTATION

4.1 Working Model

Figure 4.1: System Design

12
We have developed a Agricultural bot that uses a Web based API to implement
manual as well as automatic control.The bot is powered by rechargeable batteries
which can be swapped instantly to make sure minimum downtime

Figure 4.2: Robot Design

13
4.2 Software and Hardware Used
4.2.1 Hardware
The Robot will be equipped with two pumps for spraying on either side for maxi-
mum coverage, secondly two wheel encoders are used for calculating the distance
travelled and for making precise 90 degree turns. The bot makes use of a battery
pack which is utilized solely for hardware and no power is lost in on-board comput-
ing. 4 DC motors are being used for which are interfaced using Node MCU. The
wheels used have ridges for traction and can further be improved using a bigger
set of offroad wheels

14
1. Control Board:
Processing power is provided by an Arduino board. The board is powered
by Node MCU which is a microcontroller with 1 MB flash memory for stor-
ing the code. The microcontroller can be programmed by C language-like
“processing programming language.”

Figure 4.3: Node MCU Pinout

NodeMCU ESP8266 Specifications & Features:

• Operating Voltage: 3.3V

• Input Voltage: 7-12V

• Digital I/O Pins (DIO): 16

• Analog Input Pins (ADC): 1

15
 • UARTs: 1

• SPIs: 1

• I2Cs: 1

• Flash Memory: 4 MB

• SRAM: 64 KB

• Clock Speed: 80 MHz

• USB-TTL based on CP2102 is included onboard, Enabling Plug n Play

PCB Antenna

• Small Sized module to fit smartly inside your IoT projects

2. Lm393 Motor Speed Measuring Sensor Module:

Each wheel is equipped with a sensor which is basically a pair of infra-red

transmitter and receiver. By counting the holes in the wheel and knowing
the wheel diameter, the robot can encode the distance traveled.

Figure 4.4: IR Wheel encoder sensor

16
The LM393 is a dual independent accuracy voltage integrated circuit oper-
ated with single or else split supply. These ICs comprises two independent
voltage comparators to operate from an only power supply more than a wide
variety of voltages. Working with two supplies is also achievable as long as
the variation among the two supply voltages is 2 volts to 36 volts, & VCC
is minimum 1.5 volts extra positive than the i/p voltage. Widely used in
motor speed detection, pulse count, the position limit, etc. The DO output
interface can be directly connected to a micro-controller IO port, if there is
a block detection sensor it can also be used to determine the speed of the
motor.

Main technical characteristics:

• Dimensions: 32 x 14 x 7mm.

• The sensor reading slot has a width of 5mm

• Two outputs, one Digital and one Analog.

• LED power indicator.

• LED indicator of the output pulses of pin D0.

Features

• The working voltage of 3.3 V to 5 V.

• Output form: digital switch output (0 and 1).

• Uses the LM393 wide voltage comparator IC.

• Can provide upto 15mA drive current.

17
3. L293D (Motor Driver):

Figure 4.5: L293D Specifications

L293D is a typical Motor driver or Motor Driver IC which allows DC motor

to drive on either direction. L293D is a 16- pin IC which can control a set
of two DC motors simultaneously in any direction. It means that you can
control two DC motor with a single L293D IC. Dual H-bridge Motor Driver
integrated circuit (IC). L293D is a dual H-bridge motor driver integrated cir-
cuit (IC). Motor drivers act as current amplifiers since they take a lowcurrent
control signal and provide a higher-current signal. This higher current sig-
nal is used to drive the motors. L293D contains two inbuilt H-bridge driver
circuits. In its common mode of operation, two DC motors can be driven
simultaneously, both in forward and reverse direction. The motor operations
of two motors can be controlled by input logic at pins 2 and 7 and 10 and 15.
Input logic 00 or 11 will stop the corresponding motor. Logic 01 and 10 will
rotate it in clockwise and anticlockwise directions, respectively.Enable pins
1 and 9 (corresponding to the two motors) must be high for motors to start

18
 operating. When an enable input is high, the associated driver gets enabled.
As a result, the outputs become active and work in phase with their inputs.
Similarly, when the enable input is low, that driver is disabled, their outputs
are off and in the high-impedance state.

4. Geared Motor:

Figure 4.6: BO Motor

DC motor (BO) Battery Operation. Dc motor converts electrical energy into

mechanical energy. Why DC gear motor used in robot Motor control circuit.
... Speed of motor is counted in terms of rotations of the soft per minute is
called RPM. RPM means Revolution per Minute. The two wheels are driven
by a pair of servo motors which are interfaced to the Node MCU board
through an L293D dual Hbridge. An L293D can drive two servo motors
or two DC motors – which can be controlled in both clockwise and counter
clockwise direction. It has output current of 600 mA and peak output current
of 1.2A per channel. The in-built diodes protect the circuit from back EMF
(Electro Magnetic Force) at the outputs. Supply voltage range vary from 4.5
V to 36 V, making L293D a flexible choice for a motor driver.

19
5. Battery Pack:

Figure 4.7: Battery Specifications

We have used rechargable three 18650 Li-ion batteries providing a total of

3000 mAh to the system which a provides a total of 12V.All power electronics
works on 12V but digital circuitry uses 5V.

20
6. Mini Submersible pump motor :

Figure 4.8: Mini Submersible pump motor

We have used two DC centrifugal pumps for spraying pesticides or fertilizers

which can be controlled by the API.A submersible pump (or sub pump,
electric submersible pump (ESP)) is a device which has a hermetically sealed
motor close-coupled to the pump body. The whole assembly is submerged
in the fluid to be pumped. Submersible pumps push fluid to the surface
as opposed to jet pumps having to pull fluids. The motor is hermetically
sealed and close-coupled to the body of the pump. A submersible pump
pushes water to the surface by converting rotary energy into kinetic energy
into pressure energy. The major advantage to a submersible pump is that it
never has to be primed, because it is already submerged in the fluid.

21
7. TIP41 Darlington Pair Switch:

Figure 4.9: TIP41 Pinout

TIP41 transistor can be used as a solid state switch. If the transistor is

operated in the saturation region then it acts as closed switch and when it
is operated in the cut off region then it behaves as an open switch.

The transistor operates as a Single Pole Single Throw (SPST) solid state
switch. When a zero input signal applied to the base of the transistor, it
acts as an open switch. If a positive signal applied at the input terminal
then it acts like a closed switch.

When the transistor operating as switch, in the cut off region the current
through the transistor is zero and voltage across it is maximum, and in the
saturation region the transistor current is maximum and voltage across is
zero. Therefore, both the on – state and off – state power loss is zero in the
transistor switch.

22
 Figure 4.10: TIP41 Specifications

8. LM7805:

Figure 4.11: LM7805

LM7805 is an industry-standard linear voltage regulator. It is a member

of the 78xx devices with a constant output voltage. The xx represents the
output voltage value of devices in the 78 families.

The LM7805 is a three-terminal linear voltage regulator IC, and it works

with AC currents. It is a common component in circuits that require positive

23
voltage regulators.

The 7805 regulator comes in different versions. The TO-92 version is made of
plastic hence works better with low-power circuits. The T-O3 version comes
in an all-metal casing for easier heat sinking.
Features

• Output Current up to 1A

• Output Voltages of 5, 6, 8, 9, 10, 12, 15, 18, 24V

• Thermal Overload Protection

• Short Circuit Protection

• Output Transistor Safe Operating Area Protection

Figure 4.12: Internal Block Diagram LM7805

24
4.2.2 Software
1. API Implementation: We have exposed the bot control API through an
http port . The various commands are

• Forward

• Reverse

• Right

• Left

• Sprayboth (for activating the pumps).

Automated scripts can also interface with the bot for controlling various
above mentioned commands.

2. Wireless LAN Wireless routers have been used to expose the bot API to
specific domain name. We have interfaced our bot on ( http://agribot/ )

25
Chapter 5

RESULTS AND DISCUSSION

5.1 Top View of Bot

hgkgkjgkjhkjhk

26
Figure 5.1: Top View of Bot

27
Chapter 6

CONCLUSIONS

This exercise is a study about the ability to equip a small mobile robot with the
ability to navigate in unknown environment like agricultural fields based on our
decisions. For the robot to take actions it relies on inputs from wheel rotation
decoders and commands from the API.

6.1 Project Highlights

Advantages:
• Compact and efficient bot.
• Can easily be automated.
Limitations:
• Range and terrain.
• Limited fertilizer capacity .
Features:
1)Turning:The bot can take precise 90 degree turns using IR wheel encoders.
2) Reachability :Due to compact design the bot can reach places that are difficult
to traverse.
3) Spraying: Multiple spraying techniques can be implemented like drip, spray and
stream.

28
Chapter 7

ACHIEVEMENTS

Figure 7.1: AVISHKAR

29
30
31
REFERENCES

[1] Rahul Kumar, Peni Jitoko, Sumeet Kumar, Krishneel Pillay, Pratish Prakash,
Asneet Sagar, Ram Singh, Utkal Mehta,Maze Solving Robot with Automated
Obstacle Avoidance,Procedia Computer Science, Volume 105,
2017.https://doi.org/10.1016/j.procs.2017.01.192.

[2] S. Mishra and P. Bande, ”Maze Solving Algorithms for Micro Mouse,” 2008
IEEE International Conference on Signal Image Technology and Internet Based
Systems, 2008, pp. 86-93, doi: 10.1109/SITIS.2008.104.

[3] E. Ecsedi, H. Silaghi, E. Mihok and V. Spoială, ”The Development of an

Autonomous Maze Robot,” 2019 15th International Conference on Engineering
of Modern Electric Systems (EMES), 2019, pp. 169-172, doi:
10.1109/EMES.2019.8795166.

32
Chapter 8

EXTRAS

BT M MWT
W←W◦ B←B◦ (8.1)
BT BW BWWT
div(MjjBW) for KL updates as below

M M
BT BW BW
WT
W←W◦ B←B◦ (8.2)
BT · 1 1 · WT

M
BT (BW)2
W←W◦ 1
BT · BW

M
WT
(BW)2
B ←B◦ 1
·WT
(8.3)
BW

Introduction
Experimental Set-up

1). PRAAT [..] software was used to mono record the vowels /a/, /u/ and /i/ -
10 each from female and male speakers at 12000 samples/sec.

2). The signal in time domain was converted to Short time Fourier Transform
(STFT) representation and the magnitude is utilised. The FFT size for the
experiments was set at 1024 points. The Hanning window with a window-
size of 512 points and hop size of 256 points was engaged.

33
Figure 8.1: NMF flowchart with Mixed signal matrix - M, Basis function matrix
- B, Weight matrix - W

3). The number of iteration for Kullback-Leibler cost function and its multi-
plicative updates was set at 200 for experimental purpose.

4). The recording was done in a laboratory with other electrical appliances e.g.
air-condition, fans switched off.

34
Table 8.1: Pitch and Formants of Original and Reconstructed vowel /a/

35