Proceedings of the Second International Conference on Innovative Mechanisms for Industry Applications (ICIMIA 2020)
IEEE Xplore Part Number: CFP20K58-ART; ISBN: 978-1-7281-4167-1
Design and Development of an Agri-bot for
Automatic Seeding and Watering Applications
Kruthika Ramesh, Prajwal K T*, Roopini C, Monish Gowda M H, V V S N Sitaram Gupta,
Department of Electrical Engineering, M S Ramaiah University of Applied Sciences, Bangalore, India
*Email: prajwal.ee.et@msruas.ac.in
Abstract— Agriculture is the backbone of rural India. may not need human support. To bring complete automation
Farmers face problems such as lack of timely availability of without manual intervention in agricultre will take quite
efficient workforce, as many have migrated from country side. sometime. The aim of this research is to find out how well the
Hence, to reduce the burden of farmers, automation in the field
concept works out in the Indian context. Based on the inputs
of farming is necessary. Automated robots are being developed
over the past two decades to assist the agriculture activities like from the end user the same robot can be rebuilt according the
ploughing, sowing, weeding, pesticide spraying and fruit picking requirmental changes.
all over the world. This paper deals with building up an The robot built during the prototype phase is a semi-automated
indigenous low-cost semi-automatic robot prototype that carries machine that sows a particular seed as per its depth and
out a couple of farming processes. The robot developed in this spacing specifications. It also monitors the soil moisture
work is semi-automatic and will be able to sow seeds based on content and automatically waters the plants. This regulates
seed spacing and depth while also regulating watering of plants. water usage and reduces water wastage. Several other above-
The applications of this can range from efficient and intensive mentioned features can be added and make it more user
commercial farming to using it for research and backyard
friendly in future versions. The use case of this can be several
gardening purposes. The developed robot prototype is tested for
its functionality and performance in a restricted area. The robot ranging from cheaper but more intensive commercial farming,
is able to automatically seed and water according the path set by growing crops on a small-scale for research purposes and also
the user using the GUI that was developed. The amount of small-scale backyard garden with plants of your choice.
watering is based on the soil moisture sensor reading that is Several other functions like ploughing, weeding and pesticide
taken between the two plants The developed robot can be spraying can be carried out by these bots as an extended
extended further by mounting it on a DC motor chassis which function.
can be used to move the robot in the entire field.
Keywords— Integrated Development Environment (IDE). The overall objective is to build an indigenous low-cost semi-
Global Positioning S ystem (GPS ), Global S ystem for Mobile
automatic robot prototype that carries out a couple of farming
Communication (GS M), Graphical User Interface (GUI), Indian
National Rupee (INR), Revolutions per Minute (RPM), Parts per processes.
million (PPM).
II. LIT ERAT URE REVIEW
I. INT RODUCT ION In [1], a review of the different types of agribots which are
already available is discussed. An image processing algorithm
India is a land of agriculture. More than 70 % of the people in is used to detect the obstacles in the path of the robot arm.
India carry out farming as their primary occupation, esp ecially Different techniques to send commands to the bot carry out
in rural areas. This caters to the food demand created by the functions especially the DTMF technique for long range is
ever-growing population. The economy of the country are also discussed. The system that uses an image processing
greatly dependent on the income from it. Agriculture in India algorithm using IR sensors which may be difficult to work in
is still carried out using the old crude methods that are labour real world scenario due to different external factors.Bluetooth
intensive. This results in low yields, food shortage and low is used to establish communication.
profits. This will also directly affect the common people as
food commodity prices increase rapidly as the demand does In [2], the authors have discussed about automatic ploughing,
not match the supply. Hence, automation is required in the seeding, fertilizing and watering mechanisms. The automatic
field of agriculture which reduces the manpower to a certain seeding and fertilizing is done using solenoid. Soil moisture
extent to carry out time-consuming and repetitve jobs. sensor is used for automatic watering application using
The concept of automation in the field of agriculture is a Rasbeery pi and internet. Auduino Atmega 328 is used for bot
couple of decades old. Several robots have been developed for processes and Rasberry pi for communication with the bot.
autonomous ploughing, vegetable picking, tree climbing, internet system is used by farmers to communicate with the
farmbot, hortibot, beebot, riceplanting robot etc. that may or
978-1-7281-4167-1/20/$31.00 ©2020 IEEE 686
Authorized licensed use limited to: University of Exeter. Downloaded on May 07,2020 at 05:41:46 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the Second International Conference on Innovative Mechanisms for Industry Applications (ICIMIA 2020)
IEEE Xplore Part Number: CFP20K58-ART; ISBN: 978-1-7281-4167-1
bot to automatically carry out the processes; which reduces the a balanced solution. The functional requirements are listed
burden of the farmers. down along with hardware, software and mechanical
requirements.
In [3], purpose is to develop a fully autonomous agribot for
agricultural applications. GPS technology is used to locate The functional requirements of the system is that it should
robot in the field; Ultrasound sensor is used for obstacle perform three functions of automatic sowing, soil moisture
avoidance: LPC arm controller and battery management testing and automatic watering. To carry out this process the
system is used that drives the geared DC motors using Skid- hardware, software and mechanical requirements are jotted
steering mechanism; using encoders, sonar sensors and camera down in the following sections.
for navigation using image processing algorithms . Integration
A. Hardware Requirements
of drivers, actuators, control system, communication system,
energy management system, task management system and The following subcomponents are required to carry out the
sensors are mounted on a suitable platform according to required functions.
required criteria; integrating with cloud services using x Soil Moisture testing requires a soil moisture
TARBIL yet to be done. The built robot was tested in open measuring unit with corresponding microcontroller
fields but since environmental considerations are not met its x Motor requirement for movement in X Y Z axis –
design was changed during implementation. Stepper/DC
x Motor driver for driving motors
In[4], developing an automatic drip irrigation system based on x Mechanical Limit switch requirements to restrict
GSM was the main focus of this paper. The system uses motor movement
several sensors to obtain the status of the field and the x Encoder requirement for more precision and closed
irrigation schedule is based on this data; Micro-controller and loop
App/DTMF is used for control and communication x Wireless data transmission through Bluetooth/Wi fi
respectively. The author developed a system which plans and x Bidirectional Level converter for 3.3V to 5V
schedules the watering process based on real time information conversion
of the field. Several problems in irrigation like physical work x Solenoid for seeding and submersible DC pump for
of farmer to control irrigation, wastage of water, wastage of Watering
time is solved. x To switch on the DC motor and Solenoid, a
MOSFET driver is used
In [5], the focus is to develop a robot for all agricultural tasks x Voltage regulator for deriving required voltage for
like ploughing, seed dispensing, fruit picking and pesticide controller and driver supply from a 12V, 4A power
spraying while also providing option for manual control. The
supply (Adapter).
system unit has developed using ATmega controller.
x Resistors and capacitors (for decoupling)
Ultrasonic sensor is used for navigation in field; Outside,
x Male to female, female to female, Male to Male
communication is through Bluetooth and GSM network;
bergs; Two pin power connectors for high and low
Camera and Image processing used for fruit picking and
current; Wires of different length.
weeding The paper highlights the usage of image processing
for weeding and fruit picking; Bluetooth and GSM used for B. Software Requirements
communication for watering and s ignificant potential to x To write embedded C logic to the microcontroller
develop and use such systems. The author concludes that by chosen Keil MDK ARM UV 5 or 4 is used
investing initially a large amount of money, the agricultural
x To develop Graphical User Interface (GUI) to help
tasks can be automated to a certain extent; beneficial in the users send commands to microcontroller for different
long run.
applications Microsoft Visual Studio 2010 is used
III. SYST EM ENGINEERING A PPROACH FOR SYST EM x For designing the mechanical structure, CATIA V5
DEVELOPMENT and Autodesk Inventor for mechanical designing are
used. ADAMS is used for mechanical structure
simulation
Systems Engineering (SE) approach is used to build the agri-
bot as per specification so that a balanced system is developed. C. Mechanical Requirements
SE approach makes sure that the subsystems are properly There are three mechanical design requirements for this bot.
selected as per requirement and no unnecessary wastage in The main structure is to be conceptualized and designed
cost and time is incurred. It helps in developing the best x For the seeder mechanism, a hopper/funnel with a
system with cost, schedule and performance optimization. It slot to carry seeds and drop seeds when slot is opened
deals with identifying different subsystems available that x A solenoid is used to open the slot
constitutes concept exploration phase with divergent thinking x For seeder requirements, a hopper/funnel with s lot to
for problem solving. The concept selection then involves carry seeds and drop seeds when slot is opened
convergent thinking to arrive at the best subsystem and hence,
978-1-7281-4167-1/20/$31.00 ©2020 IEEE 687
Authorized licensed use limited to: University of Exeter. Downloaded on May 07,2020 at 05:41:46 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the Second International Conference on Innovative Mechanisms for Industry Applications (ICIMIA 2020)
IEEE Xplore Part Number: CFP20K58-ART; ISBN: 978-1-7281-4167-1
x For watering, a DC motor Submersible Pump, Pipe, coordinate set, it is required that either X or Y move
Sprinkler/Shower if required simultaneously with Z using interpolation algorithm as shown
in figure 7. This makes sure time is saved while moving to a
IV. ELECT RONIC SYST EM LEVEL BLOCK DIAGRAM particular coordinate system. The last button is to stop the
The electronic system level block diagram is given in figure 1. send when the user clicks on the Stop button.
As seen in the figure, it is seen that the PC/UI is used to send The control algorithm based on interpolation is discussed in
data sets to the control logic embedded in the controller figure 7. Several control algorithms for X Y plotter was
through the UART. Soil Moisture Sensor data can be looked into and interpolation technique seemed the best. Here,
interfaced to an ADC or GPIO pin configured as input. This is the difference between two sets of coordinates is found. Let’s
then sent to the control logic. The rotary encoder and limit say it is dX, dY and dZ for X Y and Z direction respectively.
switch readings are also obtained in the control logic. Based Only two motors move simultaneously at one time. Suppose X
on all these readings, the motor drives are excited which in and Y are moving together and number X steps is greater Y,
turn rotate the motors. This also sends power to the seeder and then speed of X will more than speed of Y as shown in
water solenoid valves to open accordingly and for a particular figure 7. X is termed as a dominant axis and Y the passive
duration. axis. Similarly, if Y steps is greater than X, speed of Y
(Dominant) will be more than X (Passive). In the same way, X
and Z or Y and Z movements are coordinated for simultaneous
movement. This is continued until the last set of coordinates is
reached.
Figure 1: Electronic System Level Block Diagram
V. SOFT WARE DEVELOPMENT AND T EST ING
The system involves three main algorithms to be developed.
The first one is the main control algorithm that is used for
complete autonomous working. This involves initializing all
Figure 2: Main control algorithm flowchart
the peripherals and waiting for data sets from the PC through
Bluetooth-UART. Based on the bytes received, the respective
task is carried out as shown in Figure 2 and Figure 3. For each
data set received the controller sends an Acknowledgement for
successful transmission. The variables used in the flowchart
are self-explanatory with key.
The second algorithm is the one written on the GUI side where
the GUI based on the button clicked by the user performs a
specific function. This is shown in figure 4. The form design
of the GUI is shown in figure 5. The GUI also allows the user
to select the COM port to send data through Bluetooth to the
microcontroller. In this example, COM 21 is the Bluetooth
Device Com port. The user has to first press on Reset Homing
that sends a data set with command no=0 to the controller. The
motors come to end positions and the end is detected using
limit switches in each direction. The next command is 1 that is
used for moving in X, Y and Z directions as per requirement
(involves both Initialization and Start Task). Figure 3: Main control algorithm flowchart (contd.)
The X and Y motors during initialization move to the origin of
the coordinate system simultaneously using interpolation as
shown in an example in figure 6. While moving to the next
978-1-7281-4167-1/20/$31.00 ©2020 IEEE 688
Authorized licensed use limited to: University of Exeter. Downloaded on May 07,2020 at 05:41:46 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the Second International Conference on Innovative Mechanisms for Industry Applications (ICIMIA 2020)
IEEE Xplore Part Number: CFP20K58-ART; ISBN: 978-1-7281-4167-1
As an example, a plot area of 12 inches (X) X 21 inches (Y)
is taken. The X dir row spacing = 2 inches and Y dir column
spacing = 4 inches. The motor first resets (all move at once)
and is detected by limit switch. Two inches offset on both X
and Y is taken in order to calculate the coordinates as in figure
6. Once calculated, the system then moves to the starting
position (2 inches offset in X and Y direction) based on
interpolation technique discussed in Figure 7. This is termed
as 0, 0. The algorithm makes sure that the system moves
according to path shown in Figure 8. Once this is completed,
reset/homing used again. Interpolation based control technique
can be used for Y and Z axis and will be embedded in the code
soon as this will save time. The pros of this interpolation
algorithm are that it is easy to implement, consumes less data
Figure 4: GUI side flowchart space and memory. The cons of the developed algorithm
include less accuracy and frequent accumulation of the errors.
VI. M ECHANICAL SUBSYST EM CONCEPT UALIZAT ION
On the mechanical aspect, an X Y Z stepper motor bench was
to be designed in order to test if the product concept works.
The concept was designed with three motors as the controller
selected could support only three motors. The design of
concept in CATIA, and the structure of as sembly is as shown
in figure 8 and 9 respectively.
Figure 5: GUI Form
Figure 8: Perspective view of the designed structure in
CATIA
Figure 6: Path followed by the bot for seed sowing
Figure 7: Interpolation based algorithm for bot movement Figure 9: Assembled Structure
978-1-7281-4167-1/20/$31.00 ©2020 IEEE 689
Authorized licensed use limited to: University of Exeter. Downloaded on May 07,2020 at 05:41:46 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the Second International Conference on Innovative Mechanisms for Industry Applications (ICIMIA 2020)
IEEE Xplore Part Number: CFP20K58-ART; ISBN: 978-1-7281-4167-1
A. Seeder Design submersible pump used for pumping water. The motor is
turned ON and OFF by the microcontroller through a
After literature review and ideation [8],[9],[10],[11],[12], MOSFET driver circuit.
it was decided that the seed dispenser should carry seeds
and dispense it when it reaches a particular coordinate. To
achieve the discussed functions , it has to have a hopper/
funnel like structure to store in the seeds and a small slot
in the tube part of the funnel. The opening and closing of
the slot is carried out using a thin plate attached to the
solenoid that is mounted on the funnel. Weight balancing
of the funnel/solenoid structure is very important. The
weight will act on the structure that has been mounted on
the Y- direction and may lead to overloading of the motor
in the said direction. Hence careful design of the seeder
design is an important task. When the required coordinate
is reached, the Z-axis motor moves down to the required
number of steps (in inches) and the pointed tip of the
Figure 11: 12-24V DC submersible pump for watering
funnel is used to dig through the soil into which the seed
application
is dropped when the solenoid is on (as the slot is opened).
The duration of this process is a second after which the
slot is closed as the solenoid goes off. Finally, The Z-axis VII. CONCLUSIONS
motor brings the seeder back up by the amount of steps it
took to go down. Figure 10 shows the drawing of the
seeder design. A metal plate for soil covering attached to The developed agri-robot sows seeds automatically by
the Z axis assembly. When the movement takes place in generating coordinates based on the data entered by the user.
STM micro electronics 32-bit Cortex M4 ARM controller is
Y or X direction, the seeds at the previous coordinate are
covered with soil. used. STM32F4 microcontroller hold the control logic to use
several peripherals interfaced to it like limit switches,
encoders, soil moisture sensor, motor drivers etc. in order to
make the system work. Watering is carried out using a
submersible 8-24V DC motor water pump and a pipe. Several
minute challenges were faced and were tried to rectify during
software and hardware debug. Biggest challenge was due to
misalignment in the mechanical structure fabrication and
several measures were taken to rectify it.The total prototype
system costing is approximately around Rs.25,000. But this
will further recude once the final product has been designed
after several iterations. The farmers’burden is greatly reduced
if this sort of automation is adopted in agriculture. They don’t
have to depend much on labourers. It also saves their time as it
is efficient, flexible and user-friendly. They can even rent such
systems in case they cannot invest on it completely. In a
nutshell, it is proved that system works as desired acocording
to the functional requirements. The product specifications are
Figure 10: Seeder Design drawings
matched to and are validated with complete system testing.
Extension of this system to be adapted in field will definitely
B. Soil Moisture sensor and Watering End-tasker
help combat the problems faced by the farmers in the field.
Design
The soil moisture sensing breakout board and sensor is VIII. FUT URE SCOPE
mounted on a Medium Density Fiberboard (MDF) board
1. Mechanical structure can be modified and rectified of
that gets fixed to the Z axis L clamp as of now. Two thick
present structure for more precise system testing
12mm MDF boards are used to give the offset required
2. Mounting this/modified structure’s concept on the chasis
between seeding points to measure soil moisture and
of a bot running using DC motors after slight
pump water.cA submersible 8-24V DC motor pump is
modification.
used for pumping water through a pipe with the pipe fixed
3. Fully automate the end-tasker attachment from the present
in between the two MDF boards . Alternate arrangements
semi-automatic concept.
can be made in future. Figure 11 shows the DC
4. Carry out a proper Soil Moisture Unit Calibration
978-1-7281-4167-1/20/$31.00 ©2020 IEEE 690
Authorized licensed use limited to: University of Exeter. Downloaded on May 07,2020 at 05:41:46 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the Second International Conference on Innovative Mechanisms for Industry Applications (ICIMIA 2020)
IEEE Xplore Part Number: CFP20K58-ART; ISBN: 978-1-7281-4167-1
5. GUI can be made user friendly. Or a phone application [3] [3] Inamdar, D., et al., (2016), “ Automated Drip Irrigation System
can be developed for easy phone interface. based on Embedded System and GSM Network”, International Journal
of Innovative Research in Computer and Communication Engineering
6. Encoders used can be of more resolution and precision so (An ISO 3297: 2007 Certified Organization) Vol. 4, Issue 5, May 2016.
that they can be used to make the system more precise [4] [4] Lalwani, A., et al., 2015, December, “ A Review: Autonomous
7. One can interface the RTC and check for soil moisture Agribot For Smart Farming”, Proceedings of 46th IRF International
content everyday at a particular time Conference, 27th December 2015, Pune, India
8. Different extensions for the system can be made like [5] [5] Praveena, R. and Srimeena, R., 2015, July. “Agricultural Robot for
automatic ploughing and seeding”, In T echnological Innovation in ICT
ploughing, weeding, pesticide spraying etc. for Agriculture and Rural Development (TIAR), 2015 IEEE (pp. 17-23).
9. Long range communication system can be used if IEEE.
bluetooth does not suffice in terms of range/distance. [6] [6] Shivaprasad, B.S. and Ravishankara, M.N., (2014.) “Design and
10. In real time, the structure needs to be rigid and mounted implementation of seeding and fertilizing agriculture robot”,
on a vehicle for farming purposes which is quite International Journal of Application or Innovation in Engineering &
Management (IJAIEM), 3(6), pp.251-255.
challenging considering the indian irregular terrain of the
[7] [7] Umarkar, S. and Karwankar, A., (2016) , April. “ Automated seed
land. sowing agribot using Arduino”, In Communication and Signal
Processing (ICCSP), 2016 International Conference on (pp. 1379-1383).
IEEE.
A CKNOWLEDGMENT [8] [8] Y. N. Kumar et al., "Automated Seed Sowing Agribot," 2019 IEEE
1st International Conference on Energy, Systems and Information
Authors would take this opportunity to thank all the members Processing (ICESIP), Chennai, India, 2019, pp. 1 -5.
of the Ramaiah University of Applied Sciences, Bangalore for [9] [9] N. S. Naik, V. V. Shete and S. R. Danve, "Precision agriculture robot
their endless support and guidance in providing enormous for seeding function," 2016 International Conference on Inventive
support to carry out this research successfully. Computation T echnologies (ICICT ), Coimbatore, 2016, pp. 1 -3.
[10] [10] G. Amer, S. M. M. Mudassir and M. A. Malik, "Design and
IX. REFERENCES operation of Wi-Fi agribot integrated system," 2015 International
Conference on Industrial Instrumentation and Control (ICIC), Pune,
2015, pp. 207-212.
[1] Durmuş, H., Güneş, E.O., Kırcı, M. and Üstündağ, B.B., 2015, July. [11] [11] R. Eaton, J. Katupitiya, K. W. Siew and B. Howarth, "Autonomous
“T he design of general purpose autonomous agricultural mobile- Farming: Modeling and Control of Agricultural Machinery in a Unified
robot:“AGROBOT””, In Agro-Geoinformatics (Agro-geoinformatics), Framework," 2008 15th International Conference on Mechatronics and
2015 Fourth International Conference on (pp. 49 -53). IEEE. Machine Vision in Practice, Auckland, 2008, pp. 499 -504.
[2] [2] Gollakota, A. and Srinivas, M.B., (2011), December. “ Agribot —a [12] R. Eaton, J. Katupitiya, K. W. Siew and K. S. Dang, "Precision Guidance
multipurpose agricultural robot”, In India Conference (INDICON), 2011 of Agricultural T ractors for Autonomous Farming," 2008 2nd Annual
Annual IEEE (pp. 1-4). IEEE. IEEE Systems Conference, Montreal, Que., 2008, pp. 1 -8.
978-1-7281-4167-1/20/$31.00 ©2020 IEEE 691
Authorized licensed use limited to: University of Exeter. Downloaded on May 07,2020 at 05:41:46 UTC from IEEE Xplore. Restrictions apply.
