IJSRD - International Journal for Scientific Research & Development| Vol.

6, Issue 03, 2018 | ISSN (online): 2321-0613

Agriculture Robotic Vehicles based Pesticides Sprayer

Jaggumantry Swapna Kumari1 Kazi Khalid Abdul Karim2 Mankari Hemant Tanaji3
Bodhgire Yogesh Uttamrao4 Momin Md. Arbi Md. Husain5
1
Assistant Professor 2,3,4,5UG Student
1,2,3,4,5
Department of Electrical Engineering
1,2,3,4,5
Annasaheb Chudaman Patil College of Engineering, Kharghar, Maharashtra, India
Abstract— This project deals with the exposition of how
robotics can be applied to various fields of agriculture. One II. REVIEW OF LITERATURE
of the most important occupations in a developing country Robotics plays a major role not only in industrial, medical,
like India is agriculture. It is very important to improve the military but also in agricultural applications [2]. The
efficiency and productivity of agriculture by replacing laborer Agricultural robotic systems are used as a tool that can enable
with intelligent machines like robots using latest a transformation of practices and the adoption of new
technologies. The project proposes a new strategy to replace technologies in field and crop management. The smart
humans in various agricultural operations like detection of machine called as “green seeker sensor” that was developed
presence of pests, spraying of pesticides, spraying of at Oklahoma state university which reads a plant’s needs and
fertilizers, etc. there by providing safety to the farmers and then applies precisely the amount of fertilizer or herbicides
precision agriculture. The developed system involves needed [3]. Some authors have worked on Autonomous
designing a prototype which uses simple cost effective Pesticide Spraying Robot for spraying pesticides in
equipment’s like microprocessors, wireless camera, various greenhouse [4-6].
motors and terminal equipment’s which is an aid to the A robotic vehicle using PIC Microcontroller to
farmers in various crop field activities. control the movement of robot with the help of transmitter, a
Key words: Pesticide Sprayer, Agriculture Vehicle, Robot, receiver and a wireless camera mounted on the top of the
Automated Pests Management, Image Analysis, Object vehicle that tracks the path taken by the robot for spraying
Detection, Object Extraction harmful pesticides has been developed [1]. This robot mainly
emphasizes on pesticide spraying by farmers from a remote
I. INTRODUCTION location without directly coming in contact with it. Robotics
Agriculture is one of the most important occupations in a Technology has been implemented for weeding, sowing and
developing country like India. The problems related to the reaping of fruits from the field [7]. Another project that
traditional farming such as lack of knowledge of using developed a multipurpose machine, which is used for digging
fertilizer/pesticides, lack of man power, etc. due to which the soil, seed sowing, and leveler to close the mud and water
farmers are leaving the farming profession. Advancement in sprayer to spray water [8]. Further, companies like WIPRO
the field of robotics has widened and its applications extend are actively working on making advanced systems and robots
from home automations, military operations along with for agriculture [9]. Such techniques can be implemented for
agriculture related activities. Automation may prove as improving the productivity and efficiency of Indian
attraction to young farmers to continue their traditional agriculture in future.
profession with improved efficiency, precision along with the
safe cultivation practices. III. DESCRIPTION OF THE PROPOSED SYSTEM COMPONENTS
To provide safety to the farmers and the main Agriculture robot vehicle consist of basic equipment’s such
objective of this work is to provide precision farming. Here, as microcontroller, transmitter & receiver, driver circuit,
Robots will be replacing laborers for the farm activities like stepper motor and wireless camera. This robot vehicle which
detection of pests, spraying of pesticides/fertilizers etc. whose navigates in between the crops based on the instructions given
operations will be automated or can be controlled by the by the farmer using joystick. The transmitted instructions will
farmer. be received at the receiver end and corresponding movement
This system involves designing a prototype using of the robot will be obtained. The instruction transmitted from
equipment’s like microprocessors, wireless camera, motors the transmitter, reception at the receiver and corresponding
and terminal equipment’s which will offer an aid to the movement of the robot is depicted as shown in the block
farmers for automation. diagram in Fig. 2.1
The rationale behind the proposed project is to bring
PIC
in the improvements in safety of farmers during the crop Transmitter Receiver Microcontroller
activities like spraying chemicals, fertilizers and pesticides.
The research projects finds its relevance in the field of
Agricultural Engineering, Electrical Engineering, Electronics Stepper Motor Driver Board Opto Coupler
Engineering, Telecommunication Engineering, Mechatronics Fig. 2.1: Block Diagram of Robotic Motion
Engineering, Environmental Engineering, Biomedical These robotic vehicles mainly consist of the following parts:
Engineering, Mechanical Engineering etc.
A. Motherboard
The motherboard consists of the PIC18 micro controller,
crystal oscillator, and filters. The PIC microcontroller is the
main controller which drives the entire robot. It gives

All rights reserved by www.ijsrd.com 24

 Agriculture Robotic Vehicles based Pesticides Sprayer
(IJSRD/Vol. 6/Issue 03/2018/006)
command signals to the driver circuit which drives the stepper precise speed control is required in various autonomous
motor to run the robot. Optocouplers are present in the systems. It is well suited for open loop control system as a
motherboard, these are optically coupled and electrically precise output can be obtained as the output depends on the
isolated with the driver circuit. This protects the count of pulses sent to the motor thereby eliminating the need
microcontroller and ensures that data leaves the of sensors or feedback system and hence it reduces the overall
microcontroller and no external data enters the cost.
microcontroller. It protects the controller from large back
E. Driver Boards
electro motive force else it may destroy the components
The signal is sent to power driver circuit which further drives
B. Transmitter & Receiver the current and provides the necessary current to drive the
The block diagram of the transmitter is shown in Fig.2.2. motor. It works based on Automatic gain control method. The
signal is sent from the driver circuit to the stepper motor. The
stepper motor works according to the pulses it receives from
the driver circuit.
This project serves to lessen the laborious work of
the farmers and also proves to be very economical and
Fig. 2.2: Block Diagram of Transmitter efficient
The transmitter consists of a carrier signal generator
(oscillator) which produces the carrier wave frequency, IV. PROPOSED METHOD FOR PEST DETECTION ON PLANTS/
modulating signal generator which produces the signal to be CROPS
modulated according to the command given by the user,
mixer which mixes the signals and amplifier which increases Detection of pest can be done more accurately using image
the power of this signal then the modulated signal is sent to processing method. Image processing is the analysis and
the transmitter. The transmitter transmits this signal. manipulation of graphical images from sources such as
The receiver board consists of IC (12D), a wireless photographs and videos. There are three main steps in image
receiver of 434 MHZ and LEDs. The receiver IC consists of processing; first is the conversion of captured images into
antenna which captures the RF signal, RF amplifier which binary values that a computer can process; second, is the
amplifies the weak signal, a tuner and decoder which image enhancement and data compression; and the third is the
demodulates the information signal of particular frequency output step that consists of the display or printing of the
from the carrier signal and performs the needed control processed image. Image processing is used in such
action. The block diagram of receiver is as in the Fig-2.3. applications as satellite weather mapping, machine vision,
and computer-based pattern recognition. Here is the block
diagram of pest detection and spraying of pesticides as shown
in fig. 2.4.

Fig. 2.3: Block Diagram of Receiver

C. Wireless Camera & A Television
A wireless camera is used for live transmission of the images
of crops and the path to the remote user (farmer). The wireless
camera is mounted on the top of the robot. The viewing angle
of the camera can also be adjusted. The RF based live video
acquisition module consists of a wireless camera and in built
transmitter. This transmitted data is received by an RF Fig. 2.4: Block Diagram of Pest Detection & Spraying of
receiver and it is viewed using an appropriate terminal device Pesticides
(television). The live streaming allows the farmer to As the images/ video captured by camera is used for
specifically guide and navigate the robot. The remote detection of the pesticides. There are some types for
manipulation of robot is possible using this facility. techniques used in image processing.
Instead of using a separate microcontroller, RF There are various types of image processing method:
transmitter, RF communication controller wireless camera 1) Image Pre-Processing
with inbuilt transmitter, joystick and a single microcontroller 2) Image Enhancement
to control the entire working of robot is used for simplicity 3) Image Segmentation
and cost efficiency. 4) Feature Extraction
5) Image Classification
D. Stepper Motor
A stepper motor is a brushless DC electric motor which A. Image Pre-Processing
rotates through a number of steps. According to the input In image preprocessing, image data recorded by sensors on a
current pulse from the controller it rotates through a fixed satellite restrain errors related to geometry and brightness
angular step. Stepper motors can be directly controlled using values of the pixels. These errors are corrected using
microprocessors, computers and programmable controllers. It appropriate mathematical models which are either definite or
is ideally suited for situations where precise positioning and statistical models.

All rights reserved by www.ijsrd.com 25

 Agriculture Robotic Vehicles based Pesticides Sprayer
(IJSRD/Vol. 6/Issue 03/2018/006)
B. Image Enhancement subdivision is carried out depends on the problem being
Image enhancement is the modification of image by changing solved, i.e., the segmentation should stop when the objects of
the pixel brightness values to improve its visual impact. interest in an application have been isolated e.g., in
Image enhancement involves a collection of techniques that autonomous air-to-ground target acquisition, suppose our
are used to improve the visual appearance of an image, or to interest lies in identifying vehicles on a road, the first step is
convert the image to a form which is better suited for human to segment the road from the image and then to segment the
or machine interpretation. contents of the road down to potential vehicles. Image
Sometimes images obtained from satellites and thresholding techniques are used for image segmentation.
conventional and digital cameras lack in contrast and After thresholding a binary image is formed where
brightness because of the limitations of imaging sub systems all object pixels have one gray level and all background pixels
and illumination conditions while capturing image. Images have another - generally the object pixels are 'black' and the
may have different types of noise. In image enhancement, the background is 'white'. The best threshold is the one that
goal is to accentuate certain image features for subsequent selects all the object pixels and maps them to 'black'. Various
analysis or for image display. Examples include contrast and approaches for the automatic selection of the threshold have
edge enhancement, pseudo-coloring, noise filtering, been proposed. Thresholding can be defined as mapping of
sharpening, and magnifying. Image enhancement is useful in the gray scale into the binary set {0, 1}: equation Where S(x,
feature extraction, image analysis and an image display. The y) is the value of the segmented image, g(x, y) is the gray
enhancement process itself does not increase the inherent level of the pixel (x, y) and T(x, y) is the threshold value at
information content in the data. It simply emphasizes certain the coordinates (x, y). In the simplest case T(x,y) is
specified image characteristics. Enhancement algorithms are coordinate independent and a constant for the whole image.
generally interactive and application dependent. Some of the It can be selected, for instance, on the basis of the gray level
enhancement techniques are: histogram. When the histogram has two pronounced maxima,
a) Contrast Stretching which reflect gray levels of object(s) and background, it is
b) Noise Filtering possible to select a single threshold for the entire image. A
c) Histogram Modification method which is based on this idea and uses a correlation
1) Contrast Stretching criterion to select the best threshold is described below.
Some images (e.g. over water bodies, deserts, dense forests, Sometimes gray level histograms have only one maximum.
snow, clouds and under hazy conditions over heterogeneous This can be caused, e.g., by inhomogeneous illumination of
regions) are homogeneous i.e., they do not have much change various regions of the image. In such case it is impossible to
in their levels. In terms of histogram representation, they are select a single thresholding value for the entire image and a
characterized as the occurrence of very narrow peaks. The local binarization technique must be applied. General
homogeneity can also be due to the incorrect illumination of methods to solve the problem of binarization of in
the scene. Ultimately the images hence obtained are not easily homogeneously illuminated images, however, are not
interpretable due to poor human perceptibility. This is available.
because there exist only a narrow range of gray-levels in the Segmentation of images involves sometimes not
image having provision for wider range of gray-levels. The only the discrimination between objects and the background,
contrast stretching methods are designed exclusively for but also separation between different regions. One method for
frequently encountered situations. Different stretching such separation is known as watershed segmentation.
techniques have been developed to stretch the narrow range
to the whole of the available dynamic range. D. Feature Extraction
2) Noise Filtering The feature extraction techniques are developed to extract
Noise Filtering is used to filter the unnecessary information features in synthetic aperture radar images. This technique
from an image. It is also used to remove various types of extracts high-level features needed in order to perform
noises from the images. Mostly this feature is interactive. classification of targets. Features are those items which
Various filters like low pass, high pass, mean, median etc., uniquely describe a target, such as size, shape, composition,
are available. location etc. Segmentation techniques are used to isolate the
3) Histogram Modification desired object from the scene so that measurements can be
Histogram has a lot of importance in image enhancement. It made on it subsequently. Quantitative measurements of
reflects the characteristics of image. By modifying the object features allow classification and description of the
histogram, image characteristics can be modified. One such image.
example is Histogram Equalization. Histogram equalization When the pre-processing and the desired level of
is a nonlinear stretch that redistributes pixel values so that segmentation has been achieved, some feature extraction
there is approximately the same number of pixels with each technique is applied to the segments to obtain features, which
value within a range. The result approximates a flat is followed by application of classification and post
histogram. Therefore, contrast is increased at the peaks and
processing techniques. It is essential to focus on the feature
lessened at the tails
extraction phase as it has an observable impact on the
C. Image Segmentation efficiency of the recognition system. Feature selection of a
Segmentation is one of the key problems in image processing. feature extraction method is the single most important factor
Image segmentation is the process that subdivides an image in achieving high recognition performance. Feature
into its constituent parts or objects. The level to which this extraction has been given as “extracting from the raw data
information that is most suitable for classification purposes,

All rights reserved by www.ijsrd.com 26

 Agriculture Robotic Vehicles based Pesticides Sprayer
(IJSRD/Vol. 6/Issue 03/2018/006)
while minimizing the within class pattern variability and VI. CONCLUSION
enhancing the between class pattern variability”. Thus, This considered work is aimed to extend the application of
selection of a suitable feature extraction technique according advanced technology in the field of agriculture. This work
to the input to be applied needs to be done with utmost care. promises to overcome precision and safety challenges in field
Taking into consideration all these factors, it becomes of agriculture and farmers respectively. It reduces tedious
essential to look at the various available techniques for work like detection of pests, spraying of pesticides/ fertilizers
feature extraction in a given domain, covering vast which will results in the improvement of productivity. Hence,
possibilities of cases. encourages many people to take up agriculture as an
E. Image Classification occupation. The user friendly agriculture robot used for
spraying pesticides is an association of all basic feasible
Image classification is the labeling of a pixel or a group of
technologies, to bring out a new and needed robot to assist
pixels based on its grey value. Classification is one of the
most often used methods of information extraction. In farmers in risk involving tasks.
Classification, usually multiple features are used for a set of
pixels i.e., many images of a particular object are needed. REFERENCES
By using this process the output images is obtained [1] Aishwarya. B. V, Archana G., C. Umayal, “Agriculture
from the captured image/video. Once this output images is Robotic Vehicle Based Pesticide Sprayer,” 2015 IEEE
found area of this image is calculated and compared with it International Conference on Technological Innovations
mean area. This calculation is done so as to minimize the use in ICT for Agriculture and Rural Development (TIAR
of amount of pesticides to be sprayed on the crops. This will 2015)J. Clerk Maxwell, A Treatise on Electricity and
help the farmer in optimizing the cost and will eventually Magnetism, 3rd ed., vol. 2. Oxford: Clarendon, 1892,
increase the productivity. pp.68-73.
Once the area is calculated the amount of pesticides [2] Syed Mutahir Mohiuddin, “Agricultural Robotics and Its
to be sprayed is determined and appropriate amount of Scope in India, International Journal of Engineering
pesticides/ fertilizers are sprayed on the crops. This project Research & Technology (IJERT), IJERTV4IS070784,
will minimize the risk associates with the health and safety of Vol. 4 Issue 07, July-2015
the farmer. [3] Emily Rutto, D. Brian Arnall, “The History of the Green
Seeker Sensor”, Oklahoma Cooperative Extension
V. EXPERIMENT RESULTS Service, PSS-2260, 2002.R. Nicole, “Title of paper with
only first word capitalized,” J. Name Stand. Abbrev., in
The accurately detection and classification of the plant
press.
disease is very important for the successful cultivation of crop
[4] Philip J. Sammons, Tomonari Furukawa and Andrew
and this can be done using image processing. This program
Bulgin, “Autonomous Pesticide Spraying Robot for use
uses various techniques to segment the disease part of the
in a Greenhouse” ARC Centre of Excellence for
plant. This program also discussed some Feature extraction
Autonomous Systems, School of Mechanical and
and classification techniques to extract the features of
Manufacturing Engineering, The University of New
infected leaf and the classification of plant diseases.
South Wales, Australia, September-2005.M. Young, The
The use of ANN methods for classification of
Technical Writer’s Handbook. Mill Valley, CA:
disease in plants such as self-organizing feature map, back
University Science, 1989.
propagation algorithm, SVMs etc. can be efficiently used.
[5] Van Li Chunlei Xia Jangmyung Lee “Vision-based Pest
From these methods, we can accurately identify and classify
Detection and Automatic Spray of Greenhouse Plant”,
various plant diseases using image processing techniques.
IEEE International Symposium on Industrial Electronics
The fig. 3.1 shows the output of the program used for
(ISlE 2009) Seoul Olympic Parktel, Seoul, Korea July 5-
detection of the pest using image processing techniques.
8, 2009
[6] Vijaykumar N Chalwa1, Shilpa S Gundagi,
“Mechatronics Based Remote Controlled Agricultural
Robot”, International Journal of Emerging Trends in
Engineering Research, International Journal of
Emerging Trends in Engineering Research, Volume 2,
No.7, July 2014.
[7] Deepika R, Mrs. K. Indra Gandhi, “AGRICULTURAL
ROBOTICS” , National conference on research
advances in communication, computation ,Electrical
Science and structures , 2015
[8] Nithin P V, Shivaprakash S, “Multipurpose agricultural
Fig. 3.1: Output of Program Used for Detection of Pest robot”, International Journal of Engineering Research,
Using Image Processing Method Volume No.5, May, 2016.
[9] http://www.wipro.com/blogs/agricultural-technology-
the-robot-farmer-has-arrived

All rights reserved by www.ijsrd.com 27