Content uploaded by Prakash Kanade
Author content
All content in this area was uploaded by Prakash Kanade on Dec 02, 2021
Content may be subject to copyright.
International Journal of Soft Computing and Engineering (IJSCE)
ISSN: 2231-2307 (Online), Volume-10 Issue-4, March 2021
1
Retrieval Number: 100.1/ijsce.D34810310421
DOI:10.35940/ijsce.D3481.0310421
Journal Website: www.ijsce.org
Published By:
Blue Eyes Intelligence Engineering
and Sciences Publication
© Copyright: All rights reserved.
Arduino Based Machine Learning and IoT Smart
Irrigation System
Prakash Kanade, Jai Prakash Prasad
Abstract: We all depend on farmers in today's world. But is
anybody aware of who the farmers rely on? They don't suffer
from various irrigation issues, such as over-irrigation, under
irrigation, underwater depletion, floods, etc. We are trying to
build a project to solve some of the problems that will help
farmers overcome the challenges. Owing to inadequate
distribution or lack of control, irrigation happens because of
waste water, chemicals, which can contribute to water
contamination. Under irrigation, only enough water is provided
to the plant, which gives low soil salinity, leading to increased
soil salinity with a consequent build-up of toxic salts in areas
with high evaporation on the soil surface. This requires either
leaching to remove these salts or a drainage system to remove the
salts. We have developed a project using IoT (Internet of Things)
and ML to solve these irrigation problems (machine learning).
The hardware consists of different sensors, such as the
temperature sensor, the humidity sensor, the pH sensor, the
raspberry pi or Arduino module controlled pressure sensor and
the bolt IOT module. Our temperature sensor will predict the
area's weather condition, through which farmers will make less
use of field water. At a regular interval, our pH sensor can sense
the pH of the soil and predict whether or not this soil needs more
water. Our main aim is to automatically build an irrigation
system and to conserve water for future purposes.
Keywords: Irrigation, Automation, LeenaBOT, Arduino, soil
sensor, Robotics
I.
INTRODUCTION
On the world, agribusiness is poverty stricken in the
economy of various nations. Development is the foundation
of the economy in spite of money related movement. The
pillar of the economy is horticulture. It adds to the public
full yield. Farming meets the substance of the
comprehensive network's food and gives a couple of
unpleasant materials to organizations. Regardless, as there
are creature blocks in green environments, there would be
an enormous loss of yields. The yield will be crushed
totally. There would be a liberal portion of ranchers'
misfortunes. It is significant to shield commonplace fields or
domains from creatures to guarantee a basic segregation
from these monetary catastrophes. To address this issue, we
will structure a framework in our proposed work to protect
the passage of creatures into the home[25]. Our point behind
the standard is to make the ranch restrictive fencing, to
safeguard a basic partition in the light of animals from
misfortunes. Such restrictive fencing.
Manuscript received on January 08, 2021.
Revised Manuscript received on January 15, 2021.
Manuscript published on March 30, 2021.
* Correspondence Author
Prakash Kanade*, Researcher, Robotics, Artificial Intelligence, IoT,
USA
Jai Prakash Prasad, Professor, Don Bosco Institute of Technology,
Bangalore, India.
© The Authors. Published by Blue Eyes Intelligence Engineering and
Sciences Publication (BEIESP). This is an open access article under the CC
BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
shields the gather from harming the yield of the yield of the
proposition by expanding gathering. The structure would not
be Perilous and inconvenient to creatures similarly as
people. The focal point of the endeavor is the utilization of
the installed framework to structure a brilliant assurance
system for home protection. Electric divider utilized in flow
procedure to shield the yields from the wild animals. Due to
high-control creatures, it is generally debilitated and not just
influences wild animals, it is likewise perilous to pet
animals and even people. The electrical divider is utilized to
protect the yields, however was utilized to observe the
animals in the flow methodology camera, which is
monetarily astonishing expense. In the framework, the sign
is accessible, however it sends the message just to the forest
official not to leave people in the farmland.
II.
LITERATURE REVIEW
The machine tracks insights concerning the sensors on the
LCD and the PC. "Muhammad (2010) [3] Proposed a basic
way to deal with "Counterfeit Neural Network Controller
Automatic Irrigation Control Problem. The proposed
framework is contrasted with the ON/OFF regulator and it is
seen that the framework dependent on the ON/OFF
Controller bombs hopelessly because of its impediments.
Then again, the technique dependent on ANN has added to
the expected execution of more grounded and more solid
force. These regulators don't require past framework
encounter and have the inalienable potential to save a ton of
assets (energy and water) from ANN-based frameworks and
can deliver advanced outcomes for all types of farming
zones. Sanjukumar (2013),[4] Proposed "Advance
Technique for Automatic Motor Pumping for Agriculture
Land Purpose Based on Soil Moisture Content" was created
and effectively actualized alongside stream sensor. The
framework's principle highlights are: shut circle
programmed water system framework, control of
temperature and water use. The client can undoubtedly set
moistness levels and update the current estimation of all
boundaries on the LCD show consistently. Later on, the
gadget will likewise coordinate other fundamental soil
boundaries, to be specific soil pH and soil electrical
conductivity[24]. S Nalini Durga (2018) proposed "Brilliant
Irrigation System Based on Soil Moisture Using Iot"
Agriculture remains the area that contributes the most
noteworthy to the GDP of India. Yet, we find that
development isn't gigantic while considering innovation that
is sent in this district. There is currently a day of colossal
improvement in developments that hugy affect various
fields, for example, farming, medical services, and so forth
In our district, agribusiness is the essential occupation.
Arduino based Machine Learning and IOT Smart Irrigation System
2
Retrieval Number: 100.1/ijsce.D34810310421
DOI:10.35940/ijsce.D3481.0310421
Journal Website: www.ijsce.org
Published By:
Blue Eyes Intelligence Engineering
and Sciences Publication
© Copyright: All rights reserved.
The significant kind of revenue for India relies upon
horticulture, so the advancement of farming is significant.
Most water system frameworks are still physically
controlled in this day and age. The customary procedures
accessible are, for example, trickle water system, sprinkler
water system, and so forth It is critical to consolidate these
strategies with IoT so we can successfully shift the
utilization of water. By having different qualities from
sensors, for example, soil dampness, water level sensors,
water quality, and so forth, IoT permits to get to data and
settle on critical dynamic cycles. Remote sensor networks
are incorporated with ZigBee in paper [6] to impart soil
dampness level and temperature esteems. Utilizing GPRS,
information is communicated to a web worker by means of a
cell organization. Utilizing graphical programming,
information following can be refined through the web.
III.
PROPOSED WORK
This framework offers a component for computerizing the
way toward getting wild animals far from farmland and
furthermore gives checking to recognize approved and
approved animals and Non-approved individual. In the event
that the utilized PIR sensors sense movement, the proprietor
of the farmland is educated regarding the interruption, we
utilize Passive Infrared Sensors (PIR) to distinguish any
human body development. This data, alongside the data put
away on the cloud, can be gotten to by the individual in
control until the message is gotten. The machine at that
point tests for the quantity of PIR sensors that have gone
HIGH in the event that it is found to be a creature, if less
sensors are high it means a more modest creature and all or
the greater part of the sensors that turn high indicated it is a
bigger creature and hence proper activity is utilized to get
them far from harming the harvests. We settle on a choice
dependent on the quantity of sensors that have gone up to
computerize the creature avert gadget talked about. On the
off chance that less sensor numbers can distinguish the
movement, the essential working hypothesis is that it
implies a creature more modest in tallness, for example, a
wild pig, deer, and so forth, and we naturally turn on the
spoiled egg splash gadget, which assists with fending the
pigs off. Likewise, if the greater part or the entirety of the
PIR sensors utilized have gone high, it is normally because
of a gigantic animal, for example, the elephant, which is
another enormous risk to such farmlands, we start the
electronic sparklers to turn ON, the uproarious commotion
that prevents the bigger animals from turning on.
Fig 1:Circuit Diagram
1.
Microcontroller:
The Arduino Mega 2560 is an ATMEGA 2560 based
microcontroller module. It has 54 computerized input/yield
pins, 16 simple information sources, 4 UARTs (equipment
sequential ports), a 16 MHz gem oscillator, a USB
association, a force jack, an ICSP header, and a reset button
(of which 15 can be utilized as PWM yields).
For projects requiring more I/O lines, more sketch memory
and more RAM, the Arduino MEGA 2560 is planned. It is
the suggested board for 3D printers and activities including
mechanical technology. This gives our ventures a lot of
room and occasions to save the Arduino stage's adaptability
and viability. Utilizing Arduino Software, the Arduino Mega
2560 is modified (IDE).
Fig 2: Microcontroller
2.
Temperature sensor (LM35):
The LM35 plan comprises of accuracy melded circuit
temperature sensors whose yield voltage is straightforwardly
tantamount to Celsius temperature.
Fig 3: Temperature sensor
3.
Soil moisture sensor:
To check the volumetric water substance of the dirt, the
earth soddenness sensor is utilized. It is utilized to screen
soil clamminess substance to screen water framework in
nursery. To recognize the element of the clamminess content
present in the field of the water framework, a soddenness
sensor is utilized. It has a module for measurement
investigation in which we can set a reference assessment.
International Journal of Soft Computing and Engineering (IJSCE)
ISSN: 2231-2307 (Online), Volume-10 Issue-4, March 2021
3
Retrieval Number: 100.1/ijsce.D34810310421
DOI:10.35940/ijsce.D3481.0310421
Journal Website: www.ijsce.org
Published By:
Blue Eyes Intelligence Engineering
and Sciences Publication
© Copyright: All rights reserved.
Fig 4: Soil moisture sensor
4.
GSM Modem:
It is an exceptional type of modem that perceives a SIM card
and runs on an adaptable overseer enrollment, nearly
equivalent to a cell phone.
Fig 5: GSM modem
5.
Humidity Sensor:
The HMTC1A2 Humidity sensor module is remembered for
the structure. This incorporates the stickiness sensor
HSS1101 and the temperature sensor LM35. It has the
qualities of consistent, high exactness, snappy reaction and
extraordinary navigate. In the arrangement moistness sensor
is used to check the tenacity obvious all around the yields.
The development of dampness is an immediate result of
water vanishing from the leaves, permitting the leaves to
recoil. So the dampness development is tried and the
sprinklers are executed to accomplish the soddenness on the
harvests. The clarification behind the autonomous utilization
of temperature sensors is that temperatures over 50 ° C can't
be dictated by this model.
Fig 6: Humidity Sensor
6.
BOLT IOT Kit:
The Bolt Cloud API offers a correspondence interface
between Bolt gadgets and any outsider system, for example,
a versatile application, web worker, python programs, and
so on The API gives natural access, following,
correspondence and utility highlights for your record
associated Bolt Devices. The Bolt Cloud API utilizes the
HTTP correspondence convention, and the HTTP GET and
HTTP POST strategies are utilized. Clients would then be
able to perform activities and recover data automatically
from Bolt gadgets utilizing conventional HTTP demands.
IV.
TECHNOLOGY USED
1. Internet of Things (IoT)
The Internet of Things (IoT) is the development of Internet
access into ordinary articles and actual gadgets. Implanted
with hardware, Internet access, and different sorts of
equipment, (for example, sensors), these gadgets can be
observed and controlled distantly and can convey and
cooperate with others over the Internet.
2. Machine Learning
AI (ML) is the observational examination of calculations
and numerical models utilized by PC frameworks to play out
a specific errand, depending rather on examples and
deduction, without utilizing unequivocal directions. It is
viewed as a man-made reasoning sub-set. To settle on
expectations or choices without being explicitly modified to
play out the assignment, AI calculations build a numerical
model dependent on example information, known as
"preparing information". In a wide scope of utilizations, for
example, email separating and PC vision, AI calculations
are utilized where it is unthinkable or unrealistic to make a
conventional calculation to play out the assignment viably.
3. VPS (Virtual Private Server)
A virtual private worker (VPS) is a virtual worker that,
despite the fact that it is introduced on an actual machine
running different working frameworks, is seen by the client
as a committed/private worker. A private virtual worker is
otherwise called a devoted virtual worker (VDS). The
possibility of a virtual private worker can be best portrayed
as a virtual machine that, much as a different actual gadget
committed to a solitary client, meets the specific
requirements of a client. The virtual committed worker
offers a similar security and usefulness as that of a common
actual gadget. An assortment of virtual private workers,
each running its own working framework, can be introduced
on a solitary actual worker.
4. Bolt library
Jolt is a GPUs-streamlined C++ layout library. For
mainstream calculations, for example, filter, lessen, change,
and sort, Bolt is intended to give superior library executions.
The C++ Standard Template Library was displayed on the
Bolt interface (STL). A ton of the Bolt APIs and
customization methodologies will be known by designers
acquainted with STL.
Arduino based Machine Learning and IOT Smart Irrigation System
4
Retrieval Number: 100.1/ijsce.D34810310421
DOI:10.35940/ijsce.D3481.0310421
Journal Website: www.ijsce.org
Published By:
Blue Eyes Intelligence Engineering
and Sciences Publication
© Copyright: All rights reserved.
V.
SIMULATION RESULTS
In our venture, we are basically utilizing AI related to IoT to
achieve the undertaking including the exchange and
appropriate correspondence of information focuses. The
accompanying calculations are utilized by us here:
Polynomial Regression
Polynomial Visualizer is a mainstream information
examination/ML calculation that assists with fitting a given
informational index with a non-direct bend. To comprehend
where other information focuses may lie, the example would
then be able to be utilized.
Anomaly Detection
Location of irregularities is the strategy for finding bizarre
articles or events that change from the norm in informational
collections.
Fig 7: Implementation View
Fig 8: Result Prediction
VI.
CONCLUSION
In the current period, the issue of yield vandalization by
wild animals has become an essential social issue. Genuine
thought and a feasible understanding are required. This
endeavor accordingly passes on an excellent social
noteworthiness as it plans to handle this issue. We have
consequently constructed a framework dependent on
brilliant inserted farmland security and reconnaissance that
is ease and devours less energy also. The fundamental
objective is to evade crop misfortunes and to shield the
locale from interlopers and wild animals that represent a
significant danger to cultivating territories. Such a
framework will assist farmers with securing their manors
and fields, set aside them from essential money related
adversities, and furthermore save them from wasteful
endeavors to protect their fields. In like manner, this
framework would help them to accomplish better gather
yields, subsequently cultivating their money related
prosperity. Contingent upon different conditions, water
resources can be utilized effectively to make the plant zone
more advantageous to achieve the prerequisites of the
interest. The ideal piece of the boundaries shifts in various
seasons and at different occasions in the customized water
structure framework. Contingent upon the particular season,
water is permitted into the yield zone. The water structure in
this manner happens more in the mid-year season, less in the
turbulent season and less in the colder time of year season.
Also, particular rules, for example, plant upgrade at different
stages and environment conditions, might be considered to
choose the water prerequisite for the yield. This will
upgrade planting, setting off the monetary progression of
about our nation. Also, the water structure framework can be
interconnected with the module for the advancement of sun-
powered goals. This will obliterate the issue of solidarity not
happening in far off regions. To lead the issues of
manageability adequacy and to fulfill the need, the water
system structure can subsequently be move to another
figuring.
REFERENCES
1. S. Ghosh, S. Sayyed, K. Wani, M. Mhatre, and H. A. Hingoliwala,
“Smart irrigation: A smart drip irrigation system using cloud,
android and data mining,” in 2016 IEEE International Conference on
Advancesin Electronics, Communication and Computer Technology
(ICAECCT), pp. 236–239, IEEE, 2016.
2. S. Vaishali, S. Suraj, G. Vignesh, S. Dhivya, and S.
Udhayakumar,“Mobile integrated smart irrigation management and
monitoring systemusingiot,” in 2017 International Conference on
Communication andSignal Processing (ICCSP), pp. 2164–2167,
IEEE, 2017.
3. S. B. Saraf and D. H. Gawali, “Iot based smart irrigation monitoring
and controlling system,” in 2017 2nd IEEE International Conference
on Recent Trends in Electronics, Information & Communication
Technology (RTEICT), pp. 815–819, IEEE, 2017.
4. A. Lage and J. C. Correa, “Weather station with cellular
communication network,” in 2015 XVI Workshop on Information
Processing and Control (RPIC), Oct 2015, pp. 1–5.
5. A.V. Bosisio and M. P. Cadeddu, “Rain detection from groundbased
radiometric measurements: Validation against rain sensor
observations,” in 2015 IEEE International Geo-science and Remote
Sensing Symposium (IGARSS), July 2015, pp. 2323–2326.
6. Pramod P. J, S. V Srikanth, Vivek N, Mahesh U Patil, Sarat Chandra
Babu N, “Intelligent Intrusion Detection System (In2DS) using
Wireless Sensor Networks”, Proceedings of the 2009 IEEE
International Conference on Networking, Sensing and Control,
Okayama, Japan, March 26-29, 2009.
7. Goldstein, A., Fink, L., Meitin, A., Bohadana, S., Lutenberg, O.,
Ravid, G., 2017. Applying machine learning on sensor data for
irrigation recommendations: revealing the agronomist’s tacit
knowledge. Precis. Agric. 19, 421–
444.https://doi.org/10.1007/s11119-017-9527-4.
8. F. Capraro, D. Patino, S. Tosetti, and C. Schugurensky, “Neural
networkbased irrigation control for precision agriculture,” in 2008
IEEE International Conference on Networking, Sensing and Control,
pp. 357–362, IEEE, 2008.
9. A. Murthy, C. Green, R. Stoleru, S. Bhunia, C. Swanson, and T.
Chaspari, “Machine learning-based irrigation control optimization,”
in Proceedings of the 6th ACM International Conference on Systems
for Energy-Efficient Buildings, Cities, and Transportation, pp. 213–
222, 2019.
International Journal of Soft Computing and Engineering (IJSCE)
ISSN: 2231-2307 (Online), Volume-10 Issue-4, March 2021
5
Retrieval Number: 100.1/ijsce.D34810310421
DOI:10.35940/ijsce.D3481.0310421
Journal Website: www.ijsce.org
Published By:
Blue Eyes Intelligence Engineering
and Sciences Publication
© Copyright: All rights reserved.
10. Y. Chang, T. Huang, and N. Huang, “A machine learning based
smart irrigation system with lora p2p networks,” in 2019 20th Asia-
Pacific Network Operations and Management Symposium
(APNOMS), pp. 1–4, 2019.
11. Tang, D.; Feng, Y.; Gong, D.; Hao, W.; Cui, N. Evaluation of
artificial intelligence models for actual crop evapotranspiration
modeling in mulched and non-mulched maize croplands. Comput.
Electron. Agric. 2018, 152, 375–384.
12. Feng, Y.; Cui, N.; Gong, D.; Zhang, Q.; Zhao, L. Evaluation of
random forests and generalized regression neural networks for daily
reference evapotranspiration modelling. Agric. Water Manag. 2017,
193, 163–173.
13. SIAM—Sistema de Información Agraria de Murcia. Available
online: http://siam.imida.es/apex/f?p= 101:1:699166260304082
(accessedon 3 April 2019).
14. San-Segundo, R.; Navarro-Hellín, H.; Torres-Sánchez, R.; Hodgins,
J.; de la Torre, F. Increasing robustness in the detection of freezing
of gait in Parkinson’s disease. Electronics 2019, 8, 119.
15. Blanco, V.; Domingo, R.; Pérez-Pastor, A.; Blaya-Ros, P.J.; Torres-
Sánchez, R. Soil and plant water indicators for deficit irrigation
management of field-grown sweet cherry trees. Agric. Water Manag.
2018, 208.
16. James, G.; Witten, D.; Hastie, T.; Tibshirani, R. An Introduction to
Statistical Learning; Springer: Berlin, Germany, 2013.
17. Vapnik, V.N. An overview of statistical learning theory. IEEE Trans.
Neural Networks 1999.
18. S.PalanivelRajan, C.Vivek, M.Paranthaman, “FeasibilityAnalysis
of Portable Electroencephalography
Based Abnormal Fatigue Detection and Tele-Surveillance System”,
International Journal of Computer Science and Information Security,
ISSN No.: 1947-5500, Vol. No.: 14, Issue: 8, pp. 711-722,2016.
19. S. Sivagamasundari, S. Janani, “Home surveillance system based on
MCU and GSM”, International journal of communications and
engineering, 2014, volume 06–no.6.
20. Padmashree S. Dhake,Sumedha S. Borde, “Embedded Surveillance
System Using PIR Sensor”, International Journal of Advanced
Technology in Engineering andScience, Volume No.02, Issue No.
03, March 2014.
21. Sudhir G. Nikhade, “Wireless Sensor Network System using
Raspberry Pi and Zigbee for Environmental Monitoring
Applications”, 2015 International Conference on Smart
Technologies and Management for Computing, Communication,
Controls, Energy and Materials (ICSTM), Vel Tech
RangarajanDr.Sagunthala R&D Institute of Science and Technology,
Chennai, T.N., India. 6 - 8 May 2015.pp. 376-381.
22. S.PalanivelRajan, R.Sukanesh, “Experimental Studies on Intelligent,
Wearable and Automated Wireless Mobile Tele-Alert System for
Continuous Cardiac Surveillance”, Journal of Applied Research and
Technology, ISSN No.: 1665–6423, Vol. No. 11, Issue No.: 1,
pp.133-143, 2013
23. S.PalanivelRajan, R.Sukanesh, “Viable Investigations and Real Time
Recitation of Enhanced ECG Based Cardiac Tele-Monitoring
System for Home-Care Applications: A Systematic Evaluation”,
Telemedicine and e-Health Journal, ISSN: 1530-5627, Online ISSN:
1556-3669, Vol. No.: 19, Issue No.: 4, pp. 278-286,2013.
24. GS Mahra, P Kumar, Chandan Solanki, DS Tomar, SK Kaushik.
Identifying grass root problems and generating sustainable solutions
through participatory rural appraisal. Indian Research Journal of
Extension Education, 2016
25. M Bala, Chandan Solanki, AT Kumar, S Tushir, R Kumar. Effect of
moisture content on some physical properties of HQPM 5 quality
protein maize (Zea mays). INDIAN JOURNAL OF
AGRICULTURAL SCIENCES, 2019
AUTHORS PROFILE
Prakash Kanade19+ Years of industrial experience in
Embedded system, Robotics, AI and IoT
Dr. Jai Prakash Prasad Professor of E&C with 19+
years of Teaching experience