2015 IEEE International Conference on Technological Innovations in ICT for Agriculture and Rural Development (TIAR 2015)

Agricultural Robot for Automatic Ploughing and

Seeding
1
Amrita Sneha.A, 2Abirami.E, 3 Ankita.A,4 Mrs.R.Praveena, 5 Mrs.R.Srimeena
1, 2, 3, 4, 5
Department of Electronics and Instrumentation Engineering
Easwari Engineering College, Chennai, Tamil Nadu, India
d7attitude@gmail.com[1], abiramiisweety@gmail.com[2], ankita6595@gmail.com[3,] praveenarmn26@gmail.com[4],
rsripriya31@gmail.com[5]

Abstract: This paper strives to develop a robot capable of less as compared to the investment. Another issue is the
performing operations like automatic ploughing, seed growing demands of the world’s population. The World Health
dispensing, fruit picking and pesticide spraying. It also Organization estimates that Earth’s population will touch 9
provides manual control when required and keeps tabs on billion in 35 years which will lead to a staggering demand in
the humidity with the help of humidity sensors .The main increase of growth of food crops. Automation is the ideal
component here is the AVR At mega microcontroller that solution to overcome all the above mentioned shortcomings by
supervises the entire process. Initially the robot tills the creating machines that perform more than one operation and
entire field and proceeds to ploughing, simultaneously automating those operations to increase yield on a large scale.
dispensing seeds side by side. The device used for navigation XUE Jinlin, XU Liming [1] published a paper on “Autonomous
is an ultrasonic sensor which continuously sends data to the Agricultural Robot and its row Guidance”, at the International
microcontroller. On the field the robot operates on Conference on Measuring Technology. The objectives of this
automated mode, but outside the field is strictly operated in paper are:
manual mode. For manual control the robot uses the  To enable the farmer to plough large areas of land in
Bluetooth pairing app as control device and helps in the minimum amount of time.
navigation of the robot outside the field.  To perform automated ploughing and simultaneous
The field is fitted with humidity sensors placed at various seeding process using Advanced Virtual RISC (AVR).
spots that continuously monitor the environment for  To provide manual control with the help of Bluetooth.
humidity levels. It checks these levels with the set point for  To measure and control humidity in the field using
humidity and alerts the farmer. The alerting mechanism is humidity sensors and water sprinkler.
GSM module that sends a text message to the farmer
informing him about the breach in set point. The farmer 1.1 Robots in Agriculture - Agrobots :
then responds via SMS to either switch on the water Agriculture is humankind’s oldest and still important economic
sprinklers or ignore the alert. The water sprinklers, if on, activity, providing the food, feeder, fiber and fuel necessary for
bring down the humidity level thus providing an ideal our survival. The current trend in agricultural robot
growing environment to crop. The concept of fruit picking development is to build more smart efficient machines that
and pesticide spraying is described under the process reduce the expense of the farmer while still providing one more
domain. Farmers today spend a lot of money on machines services and higher quality which is precisely what we have
that help them decrease labor and increase yield of crops but done in this paper. Development of a robot that can perform
the profit and efficiency are very less. Hence automation is automated ploughing and seeding operation can be manually
the ideal solution to overcome all the shortcomings by navigated by the farmer and stabilizes the humidity in the
creating machines that perform one operations and environment.
automating it to increase yield on a large scale. Robotics and automation can play a significant role in
Keywords: Agrobots, Advanced Virtual Risc (AVR),
enhancing agricultural production needs. Automation can be
Bluetooth, Humidity sensors, Machine vision system
done by man in operations such as pruning thinning and
harvesting, as well as mowing, spraying and weed removal. We
I. Introduction
can also implement with the advancement in sensors and
Farmers today spend a lot of money on machines that help
control systems that allow for optimal resource and integrated
them decrease labor work and increase yield of crops. There are
disease and pest management. Alijanobi, A.A [2] published a
various machines that are available for ploughing, harvesting,
paper titled “A set up of mobile robotic unit for Fruit
spraying pesticides etc., however these machines have to be
harvesting” at the 2010 19th International workshop. Yan Li,
manually operated to perform the required operations and
chunlei Xia, Jangmyung Lee [3] published a paper titled
moreover separate machines are used for every functions. The
"Vision based pest detection and automatic spray in green
yield and profit returns from employing this equipment are very

978-1-4799-7758-1/15/$31.00 ©2015 IEEE 17

Authorized licensed use limited to: MANIPAL INSTITUTE OF TECHNOLOGY. Downloaded on November 13,2023 at 11:15:14 UTC from IEEE Xplore. Restrictions apply.
2015 IEEE International Conference on Technological Innovations in ICT for Agriculture and Rural Development (TIAR 2015)

house plant”, page no 920-925.Once the concept of Automation automotive sector. It was after this gradual development in the
and agriculture is accepted the adoption rates will become field of agriculture began to take place. Also unlike in the
high and the costs of technology will come down. Autonomous present scenario which requires large manpower and
machines will be safer, more consistent with more efficient investment for fruit picking process we can replace it with few
plant agronomy. The robot in the fig 1 shows how a robot is numbers of this device, thus paving way for precision
used in the agricultural fields. With the help of robots, agriculture.
autonomous agricultural operations such as spraying,
mechanical weed control, fruit picking, watching the farm day 3.2 Proposed system Configuration
and night, allowing farmers to reduce the environmental The farming using this kind of device is called precision
impact, increase precision in an effective manner.The farming. The main parts of the robot are the tiller, plougher
Advantage of Automated techniques are and the wheels section. Three cameras are present for praying
 Robots can work nonstop and in hazardous pesticides and for fruit picking. The tiller is a horizontal bar
environment. with a number of jagged teeth’s fixed on it to aerate or loosen
 Robots can detect presence of diseases, weed, insect the soil bed. Cong Ming, in Ligang and Fag Bo [4] published a
infestations and other stress. paper titled “Intelligent robot Mowers: A review”, Robot,
 Due to the light weight of the robots they do not Vol. 29, no 4.From this we can say that the plougher is similar
compact the soil as large machinery does. to a tiller but has sharper and longer blades to turn over the
soil. Both the tiller and plougher are crafted of sheet metal for
the prototype model. The robot section consists of 7 motors
out of which 5 are dc motors and 2 are servo motors.
Four dc motors are attached to wheels strung on
either side such that each side is driven by two motors each.
The plougher is attached with another dc motor which aids
polar movement in the plougher. However the tiller is fitted
with more sophisticated servo motor for precise angular
Fig 1 Robot in Agricultural field movement of the tiller. The servo motor serves its optimum
purpose in the speed dispensing box where it is placed to slide
the opening exactly for few inches in order to let the seeds fall
1.2 Image Processing: in the soil. The AVR microcontroller and the other interfacing
In engineering that makes use of digital signal processors, hardware are all mounted on the robot for convenience.
image processing is one form of signal processing for which On the field navigation mechanism of the robot is
the input is an image, such as a photograph or video frame. The guided by the signals being sent out from the AVR
output of image processing may be either an image or a set of microcontroller in coordination with ultrasonic sensor placed
characteristic related to the image. Most of the image- on the robot. The ultrasonic waves are emitted and received
processing techniques treat the image as a two-dimensional continuously by the sensor which on encountering a wall or
signal and applying standard signal processing techniques to it any obstacle sends signals to the microcontroller that further
like noise rejection, gray scaling etc. An image is an array or a conveys the robot to stop, else keeping moving. They help in
matrix of square pixels that are picture elements arranged in determining how much more distance is yet to be covered by
columns and rows. It can also be defined as a two dimensional the robot, this is done with the help of echo pulse which in
function f(x, y) where x and y are spatial coordinates, and the simple terms means measuring the time taken for the pulse to
amplitude at any pair of coordinates (x, y) is called the intensity leave and return to the sensor. If the waves come across a
or gray level of the image processed at that point. boundary wall then it immediately takes 180 degree turn to
proceed to the next column of the field and so on till the last
II.RELATED WORKS: boundary all is reached. The distance to be covered is shown
on a LCD display that is mounted on the robot. In this manner
Several Projects like User friendly fuzzy logic based farm the entire field is ploughed and seeded.
automation using arduino and Lab view using x bee controller
are being undertaken. Also automatic milking systems, 3.2.1 AVR Microcontroller:
irrigation and harvesting systems, Tank farming automation The AVR has Harvard Architecture where program and data
using several meter designs are practised in most of the memory are separately placed with an 8-bit RISC single chip
western countries. microcontroller. It is the heart of the agricultural robot here. It
is one of the first microcontroller families to use on-chip flash
III.SYSTEM ARCHITECTURE memory for program storage while other controllers used
programmable ROM, EPROM or EEPROM. RISC-Reduced
3.1 Existing System Instruction Set Computing, this is a CPU designed strategy
Robotics in agriculture has been a target for agricultural based on the insight that simplified instruction can provide
aspirants for many years. The first report was published from higher performance. AVR is most appropriate for battery
the February 1934 Modern Mehanix journal. This concept tries powered appliances. The main advantages of using this
to establish crisis that a farmer could be separated from a ATmega16 are its 8-bit high performance with low power
machine’s field operation where initially there was only little consumption and that it is based on enhanced RISC
progress in the automation of Agrobots or indeed an architecture with 131 powerful instructions of which most of

18

Authorized licensed use limited to: MANIPAL INSTITUTE OF TECHNOLOGY. Downloaded on November 13,2023 at 11:15:14 UTC from IEEE Xplore. Restrictions apply.
2015 IEEE International Conference on Technological Innovations in ICT for Agriculture and Rural Development (TIAR 2015)

the instructions execute in one machine cycle. For the 3.2.5 Bluetooth:
applications, Atmega16 can work on a maximum frequency of
Here Bluetooth is used as a basic universal Remote control
16MHz.It has a programmable flash memory of 16Kb and a
for Bluetooth enabled serial devices such as Bluetooth
digital to analog comparator. The controller is interfaced with
modules connected to the microcontroller. It is a short-range
the motor, water sprinkler, camera, display, Bluetooth and gsm
wireless networking technology and is used to link (or pair)
and functions these devices.
two devices, such as smart phones and headsets, cameras and
3.2.2 DC Motors: printers, and keyboards and computers, it is sometimes called
DC Motors fall into the category of Electrical a cable-replacement technology. Both devices must support
motors that converts electrical energy into mechanical energy. Bluetooth in order to be paired, if they do, though paring the
There are several kinds of DC Motors. They work on the paring is designed to happen automatically, with little to no
principle that when a current carrying conductor is placed in a user interaction. The Bluetooth module used here is a HC-05
magnetic field, it experiences a torque and has a tendency to based on SPP support. HC-05 module is an easy to use
move which is known as the motoring action. If the direction Bluetooth SPP (Serial Port Protocol) module, designed for
of electric current in the wire is reversed then the direction of transparent wireless serial connection setup. In our idea we
the rotation is also reversed. When magnetic field and electric have made use of blue tooth Control App available on Android
field interact they produce a mechanical force which causes as controlling software more like a remote control for the
the direction of rotation of this motor to change and is given manual operation of the robot.
by Fleming’s left hand rule which states that if the index
3.2.6 LCD Display:
finger, middle finger, and thumb of your left hand are
stretched mutually perpendicular to each other and if the LCD abbreviation of Liquid Crystal Display screen is an
index finger represent the direction of the magnetic field, electronic display module which is found in a wide range of
middle finger represents the direction of the electric current applications. A 16 digit display is the very basic module and is
the then the thumb represents the direction in which the force very commonly used in most of the devices and circuits.
is experienced by the shaft of the dc motor. (1)NUMERIC LCD: displays only numbers. E.g.: old
calculators. (2)ALPHANUMERIC LCD: displays numbers
3.2.3 Servo Motor: and alphabets. E.g.: Scientific calculators. (3)GRAPHICAL
Servo motor is commonly used as an error sensing LCD: displays pictures. E.g.: mobile displays. These modules
feedback control which is used to correct the performance of a are preferred over seven segments and other multi segment
system. They are equipped with a servo mechanism for precise LEDs. The reasons are due to the LCDs being economical,
control of angular position. These motors usually have a easily programmable, having no limitation of displaying
rotation limit from 90 degree to 180 degree or to even 360 special characters or animations have comparatively better
degree. But servos do not rotate continually. Their rotation is brightness. A model of the display placed on the device is
restricted in between fixed angles. given in fig 2.
A servo motor primarily consists of a DC motor, gear
system, a position sensor which is a potentiometer, and some
of control electronics. The DC motor is connected with a gear
mechanism and provides feedback to the sensor which is the
potentiometer and from the gear box, the output of the motor
corresponds to the current position of the motor. So the change
in resistance produces an equivalent voltage from the Fig 2 LCD display
potentiometer. The pulse width modulated signal is fed
through the control wire where this pulse width is then
converted into an equivalent voltage that is compared with that 3.2.7 Ultrasonic Module HC-SR04:
of signal from the potentiometer using an error amplifier.
The HR-SR04 ultrasonic sensor uses sonar emission technique
3.2.4 Global System for Mobile Communication (GSM): to determine distance with an object just like bats or dolphins
do. A model of the HC-SR04 Ultrasonic Module is given in
A GSM/GPRS module assembles a the fig 3. It offers excellent range detection without contact but
GSM/GPRS modem with standard communication interfaces
with high accuracy of stable readings to use the package in an
like RS232 (Serial port), USB etc., so that it can be easily easy manner. Its operation is not affected by sunlight or black
interfaced with a computer or a microprocessor or material like Sharp rangefinders but acoustically soft materials
microcontroller based system. The power supply circuit is also like cloth can be difficult to detect. It comes with a complete
built in the module that can be activated using suitable ultrasonic transmitter and receiver module.
adaptor. Thus it is basically an open, digital cellular
technology used for transmitting mobile voice and data
services and supports voice calls and data transfer speeds up to
9.6 kbps, together with the transmission of SMS. Terrestrial
GSM networks now cover more than 90% of the world’s
population. A GSM satellite roaming has also extended service
access to areas where terrestrial coverage is not available. Fig 3 Ultrasonic sensor

19

Authorized licensed use limited to: MANIPAL INSTITUTE OF TECHNOLOGY. Downloaded on November 13,2023 at 11:15:14 UTC from IEEE Xplore. Restrictions apply.
2015 IEEE International Conference on Technological Innovations in ICT for Agriculture and Rural Development (TIAR 2015)

3.2.8 Humidity Sensors: The fig below describes how different parts in the robot
section are interfaced with the controller. The robot’s
Humidity is defined as the presence of water in air. The
controlling technology is the AVR microcontroller.
amount of water vapor in air can affect human comfort as well
A.Gollakota[5], Srinivas published a paper titled ”Agribot: A
as many manufacturing processes in industries. The presence
Multipurpose agricultural robot”, INDICON, vol.1, no.4,
of water vapor influences various physical, chemical and
pp.16-18(2011) which explains here that the movement of the
biological processes. In agriculture, measurement of humidity
robot is guided by the ultrasonic sensor that is fixed on the
is important for plantation protection, dew prevention, soil
robot’s body which continually sends ultrasonic waves at long
moisture monitoring etc.
distances. If these waves hits any object in their path then the
send back signals to the AVR informing the robot about the
obstacle thus prompting it to behave according to the operation
programmed in it. As the robot moves the AVR calculates the
distance yet to be covered on that strip of field and displays it
on the LCD display present in the robot.

Fig 4 humidity sensor Echo

Ultrasonic
With the development of science a chemical technology, the
Module
humidity sensor begins to use a thin piece of lithium chloride
or other semiconductor devices and measuring the resistance
which will be affected by water vapor. The resistive humidity
sensor can use this feature to gather the humidity information.
The main advantage of this sensor is that it can work in the
high temperature up to 212 degrees Fahrenheit. Fig 4 shows a Trigger
humidity sensor with its circuit connections.
3.2.9 Water sprinkler: Driving and
The type of water sprinkler used here is E-JET905F. This R Ploughing
water sprinkler comes in use when the humidity level goes X Section
below the set point. Blue
tooth
3.2.10 Camera:
Modul L293d DC
The cameras used in robots are commonly used as an image e Driver Motor
sensor. These cameras are sensitive to IR light and have an IR IC
filter placed in front of the lens. Cheap webcams may not have
such a filter, which makes them very sensitive to sunlight. T AVR
Camera calibration has always been an essential component of X MICRO
photogrammetric measurement. Self-calibration has become CONTR
essential for high-accuracy close range measurement. The OLLER
sensor that is used in the camera has a highly integrated cmos (ATmeg Tilling and
constructed array with extraction and enhancement facilities a 16) Servo Seed
built in the space. It also performs extracting any extra edges Motor Dispensin
from the image but the microcontroller has to process it. The g Section
two extreme types of calibration approaches are named as the
photogrammetric calibration and pure auto calibration
techniques. Traditionally the methods of calibration were used
to resolve a pack of non-linear equations based on triangle
measurement principle. Using servo, we can control the tilt 16 X 2
angle of the camera in the device. Displays
IV. PROCESS DESCRIPTION
The process is broadly classified into robot section and field
Fig 5 Interfacing diagram 1
section. The robot section comprises of components that take
care of the automated ploughing, seed dispensing while the
field section comprises of the humidity measurement and
The process begins with the scraping of the top surface of the
control, pesticide spraying and fruit picking operations. The
soil using the tiller fixed to the front of the robot. The entire
two sections are discussed elaborately below.
field is tilled first. The tiller loosens the soil bed enough for
4.1 Robot Section: the plougher to be driven deep into the soil in order to turn the
soil effectively. The plougher is accompanied simultaneously
seed dispensing. The tiller is controlled by a servo motor that
20

Authorized licensed use limited to: MANIPAL INSTITUTE OF TECHNOLOGY. Downloaded on November 13,2023 at 11:15:14 UTC from IEEE Xplore. Restrictions apply.
2015 IEEE International Conference on Technological Innovations in ICT for Agriculture and Rural Development (TIAR 2015)

is driven by a motor driver l293D which allows for precise communication rate is controlled using USART which is an
polar configuration in the robot’s movement. The plougher is inbuilt function of AVR.The AVR instructs the water
raised and lowered using a DC motor, which is also driven by sprinklers to either turn on or off as per the farmer’s
a L293D motor driver IC.Gwoninso [6] published a paper requirement thereby controlling humidity. Humidity is mainly
titled “Autonomous Technology”, Vol 51, No.3, 2011. The being controlled to provide an ideal environment for growth of
arrangement is such that when the robot lowers the tiller, it the crops so as to get maximum yield and produce good
raises the plougher and holds it there until the tilling process is quality crops.
completed and vice versa. DC motors are used for the rotation
of the wheels of the robot. All the above processes are
automated. For manual control, an Android phone consisting 4.2.2 Fruit picking:-
of the Bluetooth application is paired with Bluetooth on the
For the process of fruit picking we use image processing along
robot and used for navigation.
with a hydraulic arm. By the technique of image processing
the cameras locate the fruits and hydraulic arm picks the fruits.
All these action are controlled by the AVR microcontroller.

4.2 Field section:

The fig below describes how the different components on the
field section are interfaced with the AVR microcontroller.
WATER
SPRINKLER

RELAY
GS WATER
M TANK
Fig 7 Fruit picking mechanism of the robot
SI
M9 Fig 5 shows how fruit picking is done using a hydraulic arm
00 and image processing. Here we are in need to use a camera for
A AVR capturing images and we also need a fruit gripper which is an
MICRO HUMIDITY
arm tool to pick the fruit with intensive care. The position of
CONTRO SENSORS the camera is fully controllable because it is fixed to the arm
USAR
T LLER of robot and also because the camera is placed inside the
(AT mega gripper. The camera can point its optical axis at the fruit,
16) 16 X 2
INBU reducing image distortion and eliminating calibration steps
LCD
ILT that take place repetitively during apple picking. A convincing
DISPLAY advantage is that the camera is protected against collisions or
ADC
bad weather conditions, as well as against direct sunlight. The
Fig 6 Interfacing diagram 2 camera scans the tree from 40 pre-programmed positions thus
each tree is divided into 40 sectors or images.

4.2.1 Humidity measurement:-

Humidity sensors and water sprinklers are placed in fixed
spots on the field which are interfaced with an AVR controller.
A resistive type humidity sensor is used here. The AVR
controller has an inbuilt ADC that converts the analog values
coming from the humidity sensor to digital signals. An LCD
display attached to the AVR control, displays the current
humidity value along with the set point. Here, the ADC plays Fig 8 Image Processed after capture
an important role in converting the analog values to digital
values so that the LCD displays the measured values in The programmed location and orientation of the robotic arm is
readable and understandable terms. The humidity sensor stored with each image. For each sector, all ripe fruits are
continuously checks for the humidity in the environment and if identified by the image processing software, listed, and picked
it is below the set point value alerts the farmer informing him one by one in a looped task. To improve the therapeutic effect
about the compromise in set point through GSM module. The of the image processing in finding and locating the apples on
farmer instantly receives a text message informing him about the tree, the platform was designed to control the lighting
the disturbance in set point. The GSM module is interfaced to conditions to the greatest extent possible, using a canopy to
the AVR through the USART that helps perform the above cover the entire tree. The AFPM platform is also used to
function. The farmer can choose to turn on the water sprinklers reduce the effects of changing ambient lighting conditions and
or not through simple commands that can be communicated to provide a uniform background (blue) to ease locating the (red
the AVR through GSM module via text message. The
21

Authorized licensed use limited to: MANIPAL INSTITUTE OF TECHNOLOGY. Downloaded on November 13,2023 at 11:15:14 UTC from IEEE Xplore. Restrictions apply.
2015 IEEE International Conference on Technological Innovations in ICT for Agriculture and Rural Development (TIAR 2015)

and green) fruits. The output of image processing is given in upon which this area is affected. 4. To determine the color of
fig 6. Image processing is better performed on an industrial the affected area. 5. To determine size and shape of leaf, stem
PC with 2 GHz featured Pentium IV microprocessor and 1 or fruit. 6. To identify the Object correctly. Fig 8 shows how a
Gbyte featured RAM provisions running Windows XP. robot inspects the status of the leaf and an output of gray
scaling. Using the above method, the information of pest could
be calculated in detail. Through the depth of information
obtained, robot can control the arm and to how much distance
it should extend to focus the pest in the scope of the spray
nozzle. A DSP board is used to control the spray nozzle. Also
transfer of data can take place by RS-232-C fixed between
chip and computer at the robot.
Fig 9 Distance calculation to pick the fruit

During system calibration, the first step is to train the system

on the color of the fruit tree’s leaves and the blue background
canopy. A series of color thresholding steps are used to filter
out these unwanted features. First the blue is filtered out of
each of the 40 images that represent one side of the tree, and
then the leaves are filtered out. The green and the red parts of Fig 10 Processed image of the infected plant
the apples’ surfaces are located via color thresholding. After
A sprinkler similar to the water sprinkler is used to spray the
noise reduction and filling holes in the images, the fruit’s
pesticides in the field. The spraying of pesticides is done after
positions are clearly emerged .To pick one fruit, the robot has
the seeds are placed in the soil.
to determine the distance between the camera and the apple
and the path to get there. The camera measures this distance V.FUTURE WORKS
by triangulation. The measuring is done in several steps. The
camera first acquires an image, and then the camera is turned This paper is a small scale effort but the same can be
so that the fruit is situated in the center .The camera then implemented with enormous results in a large scale that
acquires a second image, and finally the diameter is calculated benefits all farmers of the world. Apart from ploughing, seed
by processing these two images. At this point, the vision dispensing, spraying pesticides and fruit picking other farming
system determines if the fruit falls within the acceptable size process like harvesting, irrigation etc. can also be implemented
range. If so, a signal is sent to the robot controller, and the in one robot thus making the machine capable of multi-
robot arm is allowed to continue to approach towards the fruit. tasking. Also looking forward to learn about and implement
While approaching the fruit, several images are processed to agricultural based agrobots like Nursery bot, Herder bot, Wine
calculate by triangulation method the remaining distance to the bot, Bee bot, and Hamster bots that would qualify the
apple while air is blowing through the gripper at a speed standards from the current precision to autonomous farming
having value about 300 m³/min to free the fruit from leaves methodologies.
that might disrupt the clarity to focus it. With each image VI.CONCLUSION
acquired, the system calculates the remaining distance using a
formula. When the selected fruit is approached, air blows An initial outcome of this study indicates that most of these
through the picking device to clear the apple from leaves. systems that which work autonomously are more flexible than
Once the apple is within a well-defined range of the gripper, traditional systems. The benefits of reduction in labor costs
the vacuum device is activated. If a vacuum around the fruit is and restrictions on the number of daily working hours
detected, the fruit is rotated and tilted softly, picked and then significantly improved. Thus it has made possible to automate
put beside. Fig 7 shows a setup to calculate distance the the most significant working routines. However some have
machine has to move in order to pick the fruit. failed due to the requirement of accuracy of specific tasks. In
addition, at this stage of development, the initial investment
4.2.3 Pesticide Spraying: and annual costs of expensive GPS system are still relatively
Camera and nozzle are installed on the end of robot arm. high but it seems possible to design economic viable robotic
When the robot moves, on finding a plant the robot stops systems for grass cutting, crop scouting and autonomous
moving and the arm inspection of the plant from bottom to top weeding. Findings show that there is a significant potential for
is done using binocular stereo vision system using one camera. applying these systems if it’s possible to impose adequate
Teruaki Mitsui, Toshiki Kagiya, Akio Inaba and Shinya Ooba control and safety regulations systems at reasonable costs.
[7] published a paper titled, “Verification of a Weeding Moreover, a comparison between different European
Robot”, Journal of Robotics and Mechatronics, Vol 20, no.2, countries indicates that labor costs, cost rotation and farm
pp.228-233, 2008.Using image processing similar to fruit structure may have a tremendous impact on the potential use
picking, i.e. (a) Original images obtained by camera (b) Gray- of these systems.
scale adjust (c) Image segmentation and area based matching VII.REFERENCES
(d) The results of fig, we can detect the weed and also the
location of it. Image analysis is usually applied for the [1] Xue Jinlin, XU Liming, “Autonomous Agriculture Robot
following purposes: 1.To detect a diseased leaf, stem or a fruit. and its row guidance”, IEEE, International Conference on
2. To spot the affected area by disease. 3. To find the extremes Measuring Technology, 2010, published

22

Authorized licensed use limited to: MANIPAL INSTITUTE OF TECHNOLOGY. Downloaded on November 13,2023 at 11:15:14 UTC from IEEE Xplore. Restrictions apply.
2015 IEEE International Conference on Technological Innovations in ICT for Agriculture and Rural Development (TIAR 2015)

[2]Aljanobi.A,”A set up of mobile robotic unit for fruit

harvesting”, IEEE International workshop, 2010, references.
[3] Yan Li, chunlei Xia, Jangmyung Lee, “Vision based pest
detection and automatic spray in green house plant”, page no
920-925, International Symposium, references.
[4] Cong Ming, in Ligang and Fag Bo Intelligent robot
Mowers: A review”, Robot, Vol. 29, no 4, 2007, in press
[5] Gwoninso, “Autonomous Technology”, Vol 51, No.3,
2011, in press.
[6] A.Gollakota, Srinivas, Agribot: A Multipurpose
agricultural robot, INDICON, vol.1, no.4, pp.16-18, 2011
[7] Teruaki Mitsui, Toshiki Kagiya, Akio Inaba and Shinya
Ooba ,”Verification of a Weeding Robot”, Journal of Robotics
and Mechatronics, Vol 20, no.2, pp.228-233,2008, published

23

Authorized licensed use limited to: MANIPAL INSTITUTE OF TECHNOLOGY. Downloaded on November 13,2023 at 11:15:14 UTC from IEEE Xplore. Restrictions apply.