2. MACHINE LEARNING
• An application of Artificial Intelligence (AI)
• Provides systems to automatically learn and
improve from experience without being
explicitly programmed
• Focuses on the development of computer
programs that can access data and use it
learn for themselves
• Primary aim- To allow the computers learn
automatically without human intervention
and adjust actions accordingly
3. EXAMPLES OF MACHINE LEARNING
The heavily hyped,
self-driving Google
car- The essence
of machine
learning
Online
recommendation
offers such as those
from Amazon and
Netflix- Machine
learning applications
for everyday life
Knowing what
customers are saying
about you on Twitter-
Machine learning
combined with
linguistic rule
creation
Fraud detection
One of the more
obvious,
important uses in
our world today
4. ADVANTAGES OF MACHINE
LEARNING
EASILY IDENTIFIES
TRENDS AND
PATTERNS
NO NEED OF
HUMAN
INTERVENTION
CONTINUOUS
IMPROVEMENT
HANDLING MULTI-
DIMENSIONAL AND
MULTI-VARIETY DATA
WIDE
APPLICATIONS
7. ROBOTICS
• A branch of
engineering that
involves the
conception design
manufacture and
operation of robots
• Robots are
programmable
machines which are
usually able to carry
out a series of actions
autonomously, or
semi-autonomously
8. REVOLUTION IN AGRICULTURE WITH
FARMING ROBOTS
Crop-
harvesting
robots
Weeding
robots
Robotic
greenhouses/
robot
farming
Aerial imagery
drones and
seed-planting
drones
9. KEY DRIVERS OF
AGRICULTURE ROBOTS
• Labour shortages decrease crop production
• High World population is expected to grow by over a
third, or 2.3 billion people, between 2009 and 2050 –
Source
• Vision and machine learning technology that allows
robots to see and train on their surroundings
• Decreased costs of connected sensors
• 40 percent of nationwide farm costs going to wages and
other labour costs
10. AGRICULTURAL ROBOTS
MARKET
Agricultural Robotics
is the logical
proliferation of
automation and cost-
effective technology
into biosystems such
as agriculture,
horticulture, and
livestock. The market
is valued at USD 3.42
billion in 2017 and is
expected to register a
CAGR of 21.1%.
11. Benefits of
Using Robotics
• User friendly, intelligent, and most
importantly affordable
• Increased their flexibility with being
capable of performing a variety of tasks
and applications
• Eliminate dangerous jobs for humans
because they are capable of working in
hazardous environments
• They can handle lifting heavy loads,
toxic substances, and repetitive tasks
• In the medical field robots are used for
intricate surgeries such as prostate
cancer surgery
12. CHALLENGES
OF ML IN
AGRICULTURE
Products like strawberry, cucumber, kiwi
are extraordinarily sensitive
Wrong calibration of robots destroys
fruits thus leads to wastage of resources
Pressure sensitivity is the start of the
problem
Huge issue- impossible for the robot to
identify which product to be plucked
13. ML IN AGRICULTURE:
APPLICATION
• Species Management
• Species Breeding- Logical
• Species Selection- Tedious process of
searching specific genes
• effectiveness of water and nutrients
usage,
• adaptation to climate change, disease
resistant;
• useful to analyse crop performance in
various climates and new
characteristics developed
16. Water management
Impacts on
• Hydrological
• Climatological
• Agronomical balance
• Estimation of daily or monthly
evapotranspiration allowing for
a more effective use of
irrigation system
• Prediction of daily due point
temperature
17. Crop Management
Yield prediction
• Yield mapping
• Estimation
• Matching of crop supply with demand
and crop management
• Incorporates computer vision
technologies to provide data related
to
• Crops
• Weather
• Economic conditions for farmers
18. Crop Quality •Accurate detection
•Classification of crop
quality
characteristics can
increase product
price and reduce
waste
•Plays a role in the
overall quality of
crops to detect
them
19. Disease Detection
• Both in open air and green house condition
• Most widely used practice in pests and
• Disease control is to uniformly spray
pesticides over the cropping area
• To be effective this approach requires
significant amount of pesticides which results
in high financial and significant environmental
costs
• ML is used as a part of general precision
agriculture management where agro
chemicals input is targeted in terms of time,
place and affected plants.
20. Weed Detection
• Most important threats to crop
production
• Biggest problem in weed fighting- difficult
to detect and discriminate from crops
• Computer vision and machine learning
can improve detection in weeds at low
cost
• No environmental issues and side effects
• In future these technologies will drive
robots which will destroy weeds
minimising the needs for herbicides
23. APPLICATION OF ROBOTICS
IN AGRICULTURE
Drones for
crop spraying
“PROSPERO” The
swarming farm-
bot
Strawberry
Harvesting
Robot
Blue River
Technology
Zea
GIGAS: Guelph
Intelligent
Greenhouse
Automation System
Driverless
Tractors
Hortibot
24. Drones for crop spraying
• Easy-to-fly devices that are designed to spray pesticides
on crops
• With drones it is also possible to capture high
resolution images of whole field for further analysis
25. Effect of usage
• Unmanned aerial vehicle (UAV) sprayer does not
need a runway
• Drones can take off and land vertically
• Flying at low altitude of several meters, the crop-
spraying can be controlled in any the sight of
distance range
• Drones are suitable for all kinds of complex
terrain, crops and plantations of varying heights
• Precise and accurate crop spraying ensures the
best coverage and application of your fertilizers or
pesticides on your lands
26. “PROSPERO” The swarming
farm-bot
• Working prototype of an
Autonomous Micro Planter
(AMP) that uses a
combination of swarm and
game theory and is the first of
four steps.
• It is meant to be deployed as
a group or "swarm".
• The other three steps involve
autonomous robots that tend
the crops, harvest them, and
finally one robot that can
plant, tend, and harvest
autonomously transitioning
from one phase to another
27. Effect of
usage
The application of the
system increases the
productivity of land on
a per unit basis
A swarm of small robots like
Prospero would have the
ability to farm inch by inch,
examining the soil before
planting each seed and
choosing the best variety for
that spot
This would maximizing the
productivity of each acre, allow
less land to be converted to farm
land, feed more people, and
provide a higher standard of living
for those people because they
would spend less of their money
on food
28. Strawberry
Harvesting Robot
Automated harvester recently
wheeled through rows of
strawberry plants here, illustrating
an emerging solution to one of the
produce industry’s most pressing
problems: a shortfall of farmhands
29. Effect of usage
Increased mechanization of the fresh-produce industry boosts
productivity, ultimately helping to tamp down price growth
It also help farmers, who are struggling with a yearlong drought in the
country’s largest produce state, get more from their fields, offsetting
higher costs
The system allows to avoid the labor shortages, existing in many
developed countries
30. Precision Lettuce Thinning by
“BLUE RIVER TECHNOLOGY”
• A lettuce-thinning robot
in California, is used for
the thinning and
weeding of lettuce to
increase yield
• Its vision system
scrutinizes each plant
and then applies
“advanced artificial
intelligence algorithms
that make plant-by-plant
decisions” to optimize
yield and then eliminate
unwanted plants
according to its
programming
31. Effect of usage
CHARACTERIZE EVERY PLOT BY COUNTING
PLANTS AND PLANT SPACING, BUILDING
CANOPY HEIGHT DISTRIBUTIONS, AND
MEASURING KEY PHYSIOLOGICAL PARAMETERS
GENERATE TABULAR DATA
AND STATISTICS FOR
EACH PLOT
BUILD PLOT IMAGE LIBRARY
THAT CONTAINS ALL IMAGES
AND PLOT RECONSTRUCTIONS
FROM EVERY PLOT
32. “Zea” Early season plant-by-plant
phenotypic measurement system
Zea delivers high-throughput,
low-hassle, high-quality plot
characterization metrics, with
a stand count error rate of
less than 5% for plants
between V2 and V5 (approx.
5-20” tall)
33. Effect of usage
Ability to do unbiased,
consistent measurements of
stand count, plant spacing,
height, greenness and leaf
area, across all of locations,
at multiple time points
Possibility to replace your
field crews and all the
hidden costs that come with
them
34. GIGAS: Guelph Intelligent Greenhouse
Automation System
GIGAS components include a vision system with multiple cameras to take images of the
plants.
At the back end is a plant database that keeps track of all the plants in the greenhouse,
with a decision support and planning element, where all the calculations are made.
Once a decision is made, that message is sent to the robot that goes and does the job
35. Effect of usage
• This robot can gently
pick an individual
beefsteak tomato, or
properly select a tomato
cluster for supper. The
robot also has a different
arm adapted for
trimming foliage and de-
leafing
• With this system it is
possible to obtain
maximum quality yield
with best timing of
harvest and other
operations
36. Driverless Tractors
Using ever-more sophisticated software coupled
with off-the-shelf technology including sensors,
radar, and GPS, the system allows an operator
working a combine to set the course of a driverless
tractor pulling a grain cart, position the cart to
receive the grain from the combine, and then send
the fully loaded cart to be unloaded
37. Effects of
usage
•Possibility to overcome the problem
of an inadequate supply of skilled
labour during planting and
harvesting
•Allow more acreage to be worked for
longer time periods
•Higher efficiency for precision
agriculture
•Smaller autonomous systems could
prove more economical, especially if
they work continuously
38. Hortibot robot for weeding
• The Hortibot is
about 3-foot-by-3-
foot, is self-
propelled, and uses
global positioning
system (GPS).
• It can recognize 25
different kinds of
weeds and
eliminate them by
using its weed-
removing
attachments.
