FARMBOT

CONTENTS

1. ABSTRACT …………………………………………………2

2. INTRODUCTION ……………………………………………3

3. OBJECTIVES ………………………………………………..4

4. HISTORY ……………………………………………………5

5. FARMBOT GENESIS ……………………………………….6

a) TOOLS ……………………………………………...8

b) SOFTWARE ………………………………………...9

 SOFTWARE CHALLAGES ………………11

c) HARDWARE………………………………………..12

6. FUTURE WORK …………………………………………….14

7. CONCLUSION ……………………………………………….15

8. REFERENCES ……………………………………………….16

Dept. of Mech. Engg., BITM, BALLARI Page 1

FARMBOT

ABSTRACT
The world’s population is growing and with that growth we must produce more
food. Due to the industrial and petrochemical revolutions, the agriculture industry has
kept up in food production, but only by compromising the soil, the environment, our
health, and the food production system itself. The increased production has largely come
from incremental changes in technology and economies of scale, but that trend is reaching
a plateau. Conventional agriculture methods are unsustainable and a paradigm shift is
needed.

Dept. of Mech. Engg., BITM, BALLARI Page 2

FARMBOT

INTRODUCTION
Agricultural automation using Farmbot is an attempt to reduce the burden
of maintaining a farm for small scale and large scale alike by automating the most
commonly performed tasks such as sowing of seeds, watering of plants and finally even
removing weeds or collecting root vegetables. In addition to all of the above the
architecture of Farmbot is modular i.e. there is a pick and place arm which is free to move
in circular and vertical direction, mounting this arm on a moving platform gives this
system freedom of movement along 3-axis.
Automation is the technique, method or system of operating or controlling a process by
highly automatic means, as by electronics devices, reducing human intervention to a
minimum.
Modular architecture refers to the design of any system composed of separate components
that can be connected together. The beauty of modular architecture is that you can replace
or add any one component (module) without affecting the rest of the system.
The area of digital electronics and intelligent systems, automation using robot has become
one of the fastest developing application-based technologies in the world. The
incorporation of the mobile communication technologies into the automation systems,
now allows the users to use mobile application on their phone to control their farms
(taken as example) from distance.

Dept. of Mech. Engg., BITM, BALLARI Page 3

FARMBOT

OBJECTIVES

The main objective of the system is to design and implement a scale model of a
cheap and open source agriculture automation system that conforms to principles of
Autonomous Robot Architectures (AuRa) mainly concentrating on modular structure,
centralized control. This system also proves as a base on which much more complex
systems might be designed.

The system uses Bluetooth for communication in this model for debugging purposes but
in real-time implementation this can be replaced by mobile GSM networks that require no
special modification on network infrastructure, or even no need for network infrastructure
at all. Here, Android mobile phone which acts as the server that bridges user and the
appliances the user wishes to control. This suitable for personal farming but in case of
large scale this can be easily replaced by a program for windows giving much more data
processing capabilities to the organization for further research and development. The
Android application is to be written in Java and the Android Development Kit SDK can
be used to rapid prototype a basic GUI that provides a convenient user interface to control
the system.

Dept. of Mech. Engg., BITM, BALLARI Page 4

FARMBOT

HISTORY

 The FarmBot project was started in 2011 when Rory Aronson whilst studying
mechanical engineering at California Polytechnic State University.
 Aronson attended and elective course in organic agriculture where he learned
about a tractor that used machine vision to detect and cover weeds which
removed the need for herbicides or manual labour , the tractor cost over $1
million USD.
 In September 2013 Aronson published a white paper outlining the goals of the
project to "Grow a community that produces free and open-source hardware
plans, software, data, and documentation enabling everyone to build and
operate a farming machine."
 The project is a response to the 60% increase food production needed due to
the growth in world population to between 7 - 9 billion by 2050 and the
potential of precision agriculture to reduce the environmental impacts of
farming by reducing water use, energy, transportation, petrochemicals and
time required to grow crops.
 In March 2014 Aronson began working on the project full-time funded by a
grant from the Shuttle worth Foundation. Firmware developer Tim Evers and
software developer Rick Carlino later joined the project as core developers
and the open source community Farmbot.cc was created to support the
development of the project.
 In March 2014 Rory Aronson created the company Farmbot.io to provide
hardware kits and software services and to serve as a funding source to
maintain the open source community. In 2014 and 2015 FarmBot was entered
into the Hackaday Prize where it became a finalist in 2015. Farmbot.io began
preorders of the first commercially available version of FarmBot, the FarmBot
Genesis, and the ninth iteration of the design in July 2016.

Dept. of Mech. Engg., BITM, BALLARI Page 5

FARMBOT

FARMBOT GENESIS

Capabilities

 The FarmBot Genesis is able to plant over 30 different crops including potatoes,
peas, squash, artichokes and chard in an area of 2.9 meters × 1.4 meters with a
maximum plant height of 0.5 meters.
 It can cultivate a variety of crops within same area at the same time and is able to
operate indoors, outdoors and covered areas.
 It is estimated that the FarmBot Genesis produces 25% fewer carbon dioxide
emissions than standard US food production

FarmBot Genesis V0.9 CAD rendering.

FarmBot Genesis V0.7 prototype working outside

Dept. of Mech. Engg., BITM, BALLARI Page 6

FARMBOT

 The Farmbot Genesis can perform almost all processes prior to harvesting
including sowing, mechanical weed control and watering.
 It requires electricity, an internet connection and water supply which be provided
using off grid solutions including a water barrel to collect rain and a solar panel
and battery to provide electricity.
 The FarmBot Genesis is able to gather data to take into account factors such as
age of the plant and local weather conditions from both local sensors and external
data from the internet.

FarmBot prototype watering for the first time

Dept. of Mech. Engg., BITM, BALLARI Page 7

FARMBOT

TOOLS

 The FarmBot Genesis performs different tasks through automatically attaching

different tools to a universal tool mount, including a seed injector, a watering
nozzle and a tool to bury weeds.
 The machine is able to weed the planted area using the weed suppressor using a
camera to identify weeds by comparing all plants in the area to the locations of
the planted seeds.

Dept. of Mech. Engg., BITM, BALLARI Page 8

FARMBOT

Software
The software for the FarmBot Genesis runs through a web interface allowing the
machine to be controlled on most internet enabled devices. The web app has the ability to
adjust different inputs including water, fertilizer and pesticide, seed spacing and
environmental factors including soil and weather conditions based on sensor readings,
location, and time of year. It is also able to build and schedule sequences by combining
and altering basic operations. The software is also able to manipulate data maps, real-time
logging and access an open plant data in the Open Farm database. All software is
available under the MIT license and is available on GitHub.

FarmBots are controlled and configured by a web application that is accessible from a
web browser on any device from any location.

FarmBot Genesis Web App on different devices.

Dept. of Mech. Engg., BITM, BALLARI Page 9

FARMBOT

FarmBot Genesis Farm Designer V7.

Dept. of Mech. Engg., BITM, BALLARI Page 10

FARMBOT

Software Challenges for the Farmbot Extensions

The design of the functional extensions aforementioned requires specific efforts from the
software engineering research community to address the underlying challenges. In this
section, we describe three of these main research challenges.

Designing resource aware software:

The design of a resource aware software requires an approach where the software will
adapt its functional and non-functional behavior dynamically according to the available
re-sources. The general problem for software engineering is therefore the automatic
optimization of resource usage for a software which requires seamless combination of
proactive and reactive adaptation of the software behavior. In the particular case of the
Farmbot system, we propose to investigate a smart resources manager that will schedule
tasks consuming these resources (network access, energy and water) based on both the
actual stocks and prevision of production. Leveraging the Farmbot access to weather
forecast and its embedded sensors, the embedded software will become aware of the
available resources and the prediction of resource productions. Taking into account a
specific model that will be built by monitoring its normal activity, the system will be able
to build a model of resource production and consumption to optimize the resource usage.

Designing domain specific optimization:

Optimizing food production and vegetable growing is not part of the expertise of a
computer scientist. Nevertheless, the software in charge of automating and optimizing the
Farmbot operations has to leverage this knowledge to both assist the user decision making
when choosing the seeds layout, and when automating the plant watering. The general
problem relates to the use of heterogeneous models and their combination in order to
optimize the operations of a software system which controls physical mechanisms (such
as Cyber Physical Systems, or Internet of Things devices). A particular instantiation of
this problem on the Farmbot system would be to leverage knowledge about multi crop
system, weather forecast, biomass growing and so on to optimize the layout of seeds and
their mutual benefit to optimize the food production.

Complex event detection through multi sensors stream analysis:

The automatic detection of mal-functions or disease is a very complex and challenging

task since it requires to crosscheck and match data coming from different sensors such as
humidity, images, and other chemical soil characteristics. Furthermore, the instantaneous
value of these sensors may not be sufficient to detect malfunctioning. The study of their
evolution over time will give significantly better results on such detection. The general
challenge for software engineering is therefore related to the ability to perform complex
relations and correlations between values produced by various sensors at different
locations and their evolution overtime. In the specific case of the Farmbot system, it
means correlating values from the evolution of the soil characteristics at various
locations, and image analysis to detect the size, holes in leafs and so on.

Dept. of Mech. Engg., BITM, BALLARI Page 11

FARMBOT

Hardware

The Farmbot Genesis is an open source hardware machine and is

designed around reproduce-ability and availability of components, it can be
created using common tools and processes meaning it is not reliant on a
single supplier.

The machine uses linear guides in the X, Y, and Z axis including the gantry
and custom flat connecting plates that can be fabricated with a number of
tools including water jet cutter, plasma cutter, laser cutter, CNC mill or
manually with a hacksaw and drill press .The FarmBot Genesis uses a belt
and pulley system including NEMA 17 stepper motors with rotary encoders,
GT2 belts and Open Builds v slot extrusions and wheels. Stainless
steel screws, t-nuts, washers, bearings, driveshaft and lead screw make the
machine weather and corrosion-resistant allowing long term operation in
outdoor environments.

Open Builds v slot wheels connecting plates, NEMA 17 stepper motor and
bolts which assemble to make the cross slide.

Farmbot is held together using 5mm aluminum plates.

Dept. of Mech. Engg., BITM, BALLARI Page 12

FARMBOT

Figure: A Farmbot prototype

Dept. of Mech. Engg., BITM, BALLARI Page 13

FARMBOT

FUTURE WORK

There is still much future scope for improving and upgrading the machine.

1. Multi-robot automated farming is yet to be implemented.

2. The passive modules need to be replaced by active modules and this can only happen

when power supply interface is designed from the place robotic interface.

3. Sensors have to be implemented to make the system more accurate.

4. The data obtained from the first generation systems have to undergo Machine Learning

to create truly intelligent systems and make them more efficient.

5. Newer modules must be designed that makes the system suitable for variety of farms.

6. Open-source frame work has to be written and successfully standardized to lead to

wide scale adoption of such modular systems.

7. Need an increase in technical know-how to operate the system.

Dept. of Mech. Engg., BITM, BALLARI Page 14

FARMBOT

CONCLUSION
The Farmbot system is a very interesting connected object which has a societal impact
through a paradigm shift for local food production. It also has many interests from a research
perspective in the software engineering community since it represents a fantastic experimental
case study for various open research problems in modern heterogeneous and distributed software
which interact directly with our physical environment.

In particular we draw in this paper several possible extensions of the Farmbot that raise
software engineering challenges, such as the design and integration of different domain-specie
languages, the composition of various kinds of models, and the computation of complex relations
and correlations from such a set of models.

Dept. of Mech. Engg., BITM, BALLARI Page 15

FARMBOT

REFERENCES
1. "Out in the Open: This FarmBot Makes Growing Food as Easy as Playing Farmville".
Retrieved 2016-08-01.

2. Jump up to:a b c d e f "FarmBot DIY agriculture robot promises to usher in the future of
farming". Retrieved 2016-08-01.

3. Jump up to:a b c d e f g h i j k l m "Details • FarmBot - Open-Source CNC Farming •

Hackaday.io". hackaday.io. Retrieved 2016-08-03.

4. Jump up to:a b c d e f g h i j "The FarmBot Genesis Brings Precision Agriculture to Your

Own Backyard". Retrieved 2016-08-02.

5. Jump up to:a b c d e "FarmBot Founder Aronson on Open Source Tech and Encouraging
Consumers to Farm - AgFunderNews". 2016-05-11. Retrieved 2016-08-02.

6. Jump up to:a b c d e says, Brian Despain (2016-06-02). "FarmBot is the world's first open
source CNC farming machine". Atmel Bits & Pieces. Retrieved 2016-08-02.

7. "The 2015 Hackaday Prize • Hackaday.io". hackaday.io. Retrieved 2016-08-23.

8. Jump up to:a b "FarmBot Genesis seeds, waters, and fertilizes your garden with life-giving
precision". Retrieved 2016-08-02.

9. Jump up to:a b "Old MacDonald Had A FarmBot, I/O, I/O, I/O". CleanTechnica.
Retrieved 2016-08-02.

10. Jump up to:a b c d e "Arduino Blog – FarmBot is an open-source CNC farming machine".
Retrieved 2016-08-02.

11. Carman, Ashley. "I bet Mesopotamian farmers didn't see FarmBot coming".
Retrieved 2016-08-02.

12. Jump up to:a b "FarmBot the Open Source Farming CNC Robot". Open Electronics.
Retrieved 2016-08-02.

13. "FarmBot Will 3D Print Your Crops and Email You When It Harvests Them".
Retrieved 2016-08-02.

Dept. of Mech. Engg., BITM, BALLARI Page 16