Use of Artificial intelligence in Animal Husbandry

1. USE OF ARTIFICIAL INTELLIGENCE IN
ANIMAL HUSBANDRY
1
Dr. Milind Nande

2. 2
C
O
N
T
E
N
T
S Introduction
What is AI
History of Artificial Intelligence
Current Global Scenario of Artificial Intelligence
Perspective of Artificial Intelligence
Application of AI in Animal Husbandry
Limitation of Artificial Intelligence
Conclusion
Strategies and Ways to Overcome Challenges
Status of Use of Artificial Intelligence in India

3. Introduction
 Global demand for various livestock products will increase up to
70% in next 30 years.
 More than half of the world population
is connected to internet through Smartphone.
 Animal Husbandry Data are widely available but they are not
used to enough to inform on production relevant task.
 In Medical Science potential Artificial Intelligence has been
developed from last four decades.
3

4. • AI is the intelligence exhibited by machine, rather than
human or animals. The intelligent agent perceive its
environment and take action to maximise success.
• It is not Man Vs Machine, however
it is synergy.
• Capability of machine to imitate
intelligent Human behaviour.
What is AI
4

5. Artificial
Intelligence
Machine
learning
Deep
Learning
• A techniques which enables machine to mimic human
behavior.
• ML is subset of AI, which use statistical methods to able
machine to improve with experience.
• DL is subset of ML, which make the computation of
multilayer neural network, that mimic the human brain .
5

6. Artificial Narrow
Intelligence (ANI)
Artificial General
Intelligence (AGI)
Artificial Super
Intelligence (ASI)
Also know as Weak AI.
It Involve applying AI
only to Specific task.
Ex: Alexa, Robtic
Milking Machine
Also Known as Strong
AI, It involve machine
that process the ability
to perform any
Intelligent task that
human can do.
Ex.: Self- driving car
It is a term referring to
the time when
capability of computer
will surpass human.
6

7.  Can Machine Think ?
1950 :‘Alan Turning’ published a landmark paper about possibility of
creating machine thinking.
1956 : The Birth of AI - John Mccarthy coined the term “ Artificial
Intelligence” at Dartmouth Conference.
1960 : General Motor Developed Ist Industrial Robot to perform
automated die.
1973:Stanford University Introduced Experimental AI in Medicine.
CASNET Model – consultation programme,
MyCIN – provided list of potential bacterial pathogen and then
recommended Ab treatment.
History of Use of Artificial Intelligence
7

8. 2007: IBM Created Open Domain Questioning Answering System
‘WATSON’ won first prize in television game show ‘Jeopardy’ .
2011: Apple developed ‘Siri and Amazon Developed Alska a natural
language processor.
2015: DL applied to image processing that stimulate the behavior of
interconnected neuron of the human brain.
8

9. Global Scenario of Artificial Intelligence
9

10. AI LANDSCAPE : GLOBALLY
85% of AI/ML World’s Start-up's are based in just 10 countries
 NITI Ayog working on to
develop National AI Strategy
Document to define road map for
India.
 Karnataka & Telangana Government
invested in creating Centre of
Excellence in Computer Science in
Bangaluru and Hyderabad.
 Digital India (2017)
10

11. 1.39 billion population
Urbanization of 35.2%
1.10 billon mobile communication.
79.0% of Total population
624 million internet user.
45.0 % of Total population
06 hr. 36 Minutes of average daily time
spend on use of internet
11

12. A global study on 6,000 adults in the United States regarding consumer’s
understanding of the new technology.
Sr.
No
Respondent’s view Response
1 Understand the term
Artificial Intelligence.
72%
2 Think they use
technology with AI.
33%
3 Actually use an AI-
powered service or
device.
77%
Reality About Artificial Intelligence?
Sr.
No
Respondent’s view Response
1 Didn’t know that AI can solve
problem.
50%
2 Didn’t knew that AI is behind
the technology used in Google
Home and Amazon Alexa.
59%
3 Open businesses using AI. if it
makes life easier. Even if they
don’t understand.
73%
PERCEPTION V/s REALITY
12

13. Conduct research on 100 respondents on AI in Agriculture by asking
Questions on Implementation and Compatibility
AI technology can revolutionize the
traditional farming.
98%
2% YES NO
75%
21%
4%
Yes No Maybe
AI based smart farming can genuinely
help farmers to acquire better yield.
13

14. Perspective of artificial Intelligence
Moore’s Law
IA (Intelligent Augmentation)
Cloud Computing
Machine Learning
14

15. 1. Moore’s Law
• In 1965, Gordon E. Moore,
the co-founder of Intel, made
this observation that became
known as Moore's Law.
• The number of transistors on
a microchip doubles every
two years, though the cost of
computers is halved.
15

16. 2. Intelligent Augmentation
• IA’s concepts revolve around
augmenting human intelligence.
• Creation of knowledge bases, image
processing tools, natural language
tools.
• Virtual Assistants: Google’s: Google
Assistant
16

17. 3. Cloud Computing
Sensor
(Capture)
Big Data
(Processing)
ML
(Analyse)
Algorithm
(Notifying)
17

18. A. Electronic Sensor (Capturing)
• A sensor is a device that detects minute changes in
its environment and sends the information to
a computer processor.
• Measuring Physical, Chemical, Biological
and Environmental Parameters.
Types of sensor
1. Temperature Sensor
2. IR Sensor
3. Ultrasonic Sensor
4. Touch Sensor
5. Light Sensor
6. Gas Sensor
18

19. B. Big data (Processing)
• It is the on-demand
availability of computer
system resources, especially
data storage and
computing power, without
direct active management
by the user.
• Google cloud – It is a suite
of public cloud services
offered by Google.
19

20. 4. Machine Learning (Analyzing)
• Machine learning is the study of
computer algorithms.
• It improve automatically through
experience and by the use of
data.
• Ex: Sensor base technology at
animal farm
Machine learning use in:.
– Email Spam and Malware
Filtering.
– Online Customer Support.
– Search Engine Result Refining.
– Product identification at Amazon
20

21. 21

22. Condition Algorithms Parameter detected
Mastitis Bag of Words (BoW), Gradient
Boosted Tree (GBT)
Somatic Cell count, Electrical
conductivity
Lameness Fog Computing, Classification
and regression Tree (CART)
Leg Movement, Neck Movement
and Image /video data
Postpartum Disease Random Forest Algorithm
(RFA)
Lactose yield, protein production
and milk yield
Algorithm use for prediction of disease condition in AH
22

23. Artificial Intelligence Use in Animal Husbandry
1. Milking Machine
2. Automated Feed
Pusher
3. Manure Removal
Robot
1. Animal Tracking System
2. Rumination Sensor
3. Activity Sensor
4. Electronic Ear Tag
5. Neck Collar Sensor
6. Camera Monitoring
7. Concentrate Feeding
Station
1. Disease detection
2. Oestrus detection
3. Data transfer into
herd management
system.
4. Concentrate feed
allocation depending
on milk yield
Machine Learning
Electronic Sensor
Electronic Data Processing
23

24. Automatic Milking System
• The Automatic Milking Machines have
Cups with Sensors that can be
attached individually to cows’ teats.
• The machines can also automatically
Clean and Sanitize the teats.
• Machines can also identify Colour,
Impurities, and Quality of milk.
• If the milk is not fit for human
consumption, it is diverted to a
separate container.
24

25. Automatic robot feeder
 Automatic robot feeder used
to feed concentrate of mixture
of roughages as per need of
farm animal.
 A scraper robot used to clean slatted
floors.
 It pushes and scrapes tirelessly, easily
traversing long passage and ensure
clean, slurry-free surfaces.
25

26. survey was conducted on 1355 dairy farmer’s to identify the interest
and concerns dairy farms when they transitioning from
conventional milking systems to AMS
Sr. No Concern Responses
1 Involved in dairy farming for > 20 yr. 74%
2 Expressed interest in transitioning to an
AMS
38%
3 A logistic regression showed that Higher Education and Larger Herd-
size influenced (P < 0.05) interest in AMS,
26

27. A survey among 107 farmers Recently invested in an automatic
milking (AM) system
Sr. No. Respondents view Percentage
1 Most farmers see themselves as a fine tune AM 43%
2 Farmers claim to be labour savers. 64.3%
3 Why they have installed an AM-system
• Social cause 67.3%
• Economic cause 32.7%
4 Implications on leisure and quality of life
• agree to spend more time with their family 86%
• quality of life of their family has improved. 75%
27

28. Drones
• Farmers Kept a manual vigilance.
whenever the livestock moves out of the
farm for grazing.
• Drones can keep track of the cattle and
herd them back from fields to barns.
• Drones can also capture the pictures of
pasture areas.
• Some companies such as TRITHI
Robotics, Dronitech, Sagar Defence
Engineering, DJI
Enterprise and Sunbirds have made
Initiative in building commercial drones in
india.
28

29. Smart Precision Farming
It is use to measure Physiological, Behavioural and Production indicators of individual
animals to improve management strategies, profitability and farm performance.
Use of PDF technologies that makes farmers less
dependent on human labour, supports them in their
(daily) management, and helps them to improve their
farm profitability
PDF monitor health and production and translate the
results in useful information.
29

30. Walking Activity
Pedometer allows monitoring of both Walking and Milking activity in dairy
farm.
Used to observe daily movements, including
milking, eating, standing, and lying, and can
detect changes in this measurement of activity.
Pedometer predicted lameness earlier than the
appearance of the clinical signs. By correlating
pedometric activity (PA) with clinical cases of
lameness.
30

31. Milk yield and Milk Electrical Conductivity
Significant changes in milk yield can be observed as early as 10 days
before diagnosis of an adverse health effect.
Electrical conductivity along with other information (e.g. milk yield, milk
flow, number of incomplete milking) may increase accuracy of detection
mastitis.
31

32. Feeding Behaviour and Intake
Found that healthy animals spent more time at the
feeder than morbid animals, and a greater percentage
of healthy animals visited to feeder immediately after
feed delivery.
Found that changes in short-term feeding behaviour of
dairy cows occurred with the onset of the health
disorders like ketosis, acute locomotory problems, and
chronic lameness.
Observation of feed intake and production by monitoring activity at the
feeder.
32

33. Motion Activity Instrumentation
Instrumentation like Electronic collar, Ear Tag, Pedometer that
consists of sensors that shows variables related to the status of
animal.
Revealed that, Animal Status is estimated by
the history of recent time of position, activity,
temperature, live weight and other
physiological parameter of all individuals in the
herd.
33

34. Animal Behavior Sensors:
 Sensors are necessary for a detailed record of
behavior of animals.
Sensors measuring
 Head Angle & Head Acceleration
 Leg Acceleration & Steps
 Swallowing & Jaw Movements,
 Biting and Chewing
 Sounds, Weight, Heart rate, Core
Temperature etc.
34

35. 35
Rumen pH and Rumen Temperature
 Continuous monitoring of ruminal pH is
possible through wireless telemetry which
has the capacity to accurately detect
subacute ruminal acidosis
 Rumen sensors to measure
Temperature, pressure/motility and pH in
rumen.
 Measurement of Ruminal pH is a reliable and
accurate diagnostic test for Ruminal Acidosis

36. Body Temperature
 The largest potential benefit of employing an
automatic body temperature monitoring system
in a dairy farm would be in early detection of
disease, illnesses, or disorders .
 body temperature is the first and foremost sign
to be detected during any disorders.
36

37. 37
Estrus Detection System
 3D-accelerometers are used to detect estrus
and calving.
 The detection of cows in heat has become more
and more difficult due to changes in animal
behaviour and management.

38. Farm Management System
 Farmers can track, monitor and manage
Health, Nutrition, Behaviour, Milking
Frequency, Milk Production and activity
level in real-time.
 Smart animal trackers can be implanted
in the cattle’s Ears, Tail, legs, Neck or any
part of the body.
 Smart cattle health tracking devices are :
Herdman, SmaXtec, Moocall, Smartbow.
38

39. Geographic Information System (GIS)
 GIS integrates hardware, software, and data for
capturing, managing, analyzing and displaying
all forms of geographically referenced
information.
 reported that GIS predict the possibility of
transmission of infectious diseases between
herds.
 Surveillance and monitoring studies, identification and location of
environmental risk factors for disease prediction, prevention and control.
39

40. Developed by NDDB, Anand
•Android tablet-based field IT application.
•Capturing of real time reliable data on
breeding, nutrition and health services
delivered at farmer’s doorstep.
•Send messages to farmers, providing
appropriate advice regarding their animals.
•Workout report are available to the
managerial team and other decision
makers for analysis.
National Livestock Identification Scheme
40
Current Status of Use of Artificial Intelligence in
Dairy Farming of India

41. Private Dairy Farm
 Chitale Dairy in Pune.
 Mukhiya Dairy, Kanodar, Palanpur, Gujrat.
 Lakshya Dairy in Haryana.
 Kopordem Farm at Valpoi in North Goa.
•Mumbai Veterinary College did research on Herdsman Software
41
 Adopted RFID based animal identification and farm
automation management system.
 RFID based animal identification, data recording
and complete farm management system applicable
for small holding dairy farmers at low cost.

42. • Adoption rate depend on various factors like farmers Age, Level of formal
Education, Farm size, types of production and overall expenses' on information and
use of technology.
• Small herd size and heavy investment in technology increase the cost of input .
• Poor availability of AI tools and computer illiteracy of farmers contribute to non-
adoption of technology.
• Unavailability of local technical expertise for interpretation and decision making.
• Different skills will be required the farmers adapt AI technologies.
• Farmers have uncertainty regarding investment in technologies due to a lack of
information.
• Lack of success stories, demonstrated effects,. leading to reduction in the interest
of the farmers to adopt the technology.
Limitation of Artificial Intelligence in AH
42

43. Strategies and Ways to Overcome Challenges
• Creation of Multidisciplinary Teams
involving scientists in various fields, like
dairying, engineers, manufacturers and
economists to study the overall scope of
artificial intelligence.
• For costly AI tools, formation of farmers’
cooperatives, self help groups or
community organizations would be one
of the solutions.
43

44. • Provide complete technical backup support to
the farmers to develop a models, which can be
replicated on a larger scale.
• Promote the technology for progressive
farmers who have sufficient risk bearing
capacity since the technology requires capital
investment.
• Effective coordination among the public,
private sectors for essential for implementing
new strategies to achieve fruitful success.
44

45. Demonstration
 Method Demonstration: it is single practice
demonstration and use to show the technique
of doing things or carry out new practices.
Result Demonstration: this method is based on
seeing is beliving. this method is use to show
the superiority of practices
Creating Awareness through Right Extension Approaches
45

46. Training Programme
• It is a process by which individuals are
helped to acquire certain specific skills
related to a given set of operations in
certain specific context only.
• In Study tour : A group of interested
person accompanied and guided by
one or more extension agents moves
out their neighbourhood to study an
learn significant improvement in farm.
46

47. Mass Media
• Mass media means the technology that is
intended to reach a mass audience.
• Current commercially AI used
Companies in dairy industry
47

48. Conclusion
• Artificial Intelligence in Animal husbandry is in its Initial Stage but it has
tremendous opportunities for improvements in individual animal and herd
management on dairy farms.
• Progressive farmers and professional can adopt AI on a limited scale to
shows potential for raising yields and economic returns on fields with
significant variability.
• Additional research needs to be undertaken to examine the adoption
process for not only successful adoption of technology but also to solve
the issues associated with the technology adoption.
48

49. 49

50. 50