An Autonomous Institution Affiliated to Visvesvaraya Technological University.

Approved by
UGC/AICTE/Govt. of Karnataka, Accredited by NAAC (Grade 'A+') Bangalore-560064, Karnataka, INDIA

AICTE-VTU Joint Teachers Training Programme for AICTE Approved & VTU
Affiliated Technical Institution Teachers One Week Teachers Training Programme
On
“ROBOTICS & ARTIFICIAL INTELLIGENCE”
Presentation On

An Overview on Practical Design,

Control and Application of AI to
Robots
Presented By
Prashanth N
Head-Centre for Robotics Research
Dept of Mechanical Engg, NMIT
 AGENDA

Introduction to Robotics

Classification of Robots

Analogous System

Bio Mimicry of Locomotion

Example: Android Robot Design

Practical Aspects of Robot Design & Control

Application of AI in Factories and Robots

Overview of Machine Learning & Its Steps

Deep Learning

Artificial Neural Networks

Expert System

Centre for Robotics Research, NMIT 2

 INTRODUCTION TO PRACTICAL ROBOT DESIGN

ORIGIN → derived from the Czech word “Robota” which literally means forced labourer or worker.

General Definition
▪ Programmable manipulator able to execute multiple operations (e.g., material and part handling), following programmed paths
to fulfil a large variety of tasks.

Fixed Robot (Industrial Manipulators)

▪ “A reprogrammable, multifunctional manipulator designed to move material, parts, tools, or specialized
devices through various programmed motions for the performance of a variety of tasks.”

Mobile Robot / Mobile Platform

▪ Reprogrammable, Multifunctional Platform with a large mobility within its environment (air, land, underwater)
▪ A system with the following functional characteristics:
• Mobility: total mobility relative to the environment
• A certain level of autonomy: limited human interaction
• Perception ability: sensing and reacting in the environment

Centre for Robotics Research, NMIT 3

 APPLICATIONS OF ROBOTS

Centre for Robotics Research, NMIT 4

 CLASSIFICATION OF ROBOTS
ROBOTS
FIXED MOBILE
Locomotive
Based on how Operating
Industrial Manipulators it Moves Environment
Drive
Mechanism

Cylindrical Spherical Cartesian Articulated SCARA

Centre for Robotics Research, NMIT 5

 ROBOT-ANALOGOUS SYSTEM
❖ Mechanisms & Structure: These include various mechanisms to transmit & transform the motion from prime
movers to the driven elements. These also consist Structures which will enable the robots to support its own
weight & as well as the payload.
• Example: Link & Joints, Gears, Belt drives etc.

Analogy

Bones & Joints Mechanical Linkages

Centre for Robotics Research, NMIT 6

 ROBOT-ANALOGOUS SYSTEM
❖ Actuators to Drive End effectors: These are the muscles of the robot which provide motion to various elements.
• Example: Electrical Motors (DC/AC) Servo Motors, Stepper Motors, Hydraulic & Pneumatic Actuators etc

Analogy

Muscles Actuators

Centre for Robotics Research, NMIT 7

 ROBOT-ANALOGOUS SYSTEM
❖ Sensors: These will provide both external & internal information to the Robot.
• These may Include, Vision Sensors (Eg: Camera), Touch Sensor (Eg: A Mechanical Switch), Proximity Sensor
(Eg: Ultrasonic, Capacitive, and Inductive Sensor etc), Light Sensors (Eg: Phototransistors, Photo resistors etc)
and Sound Sensor (Eg; Microphones) etc.

Analogy

5 Human Senses Robot Sensors

Source: http://www.ros.org/news/2011/03/sensors-supported-by-ros.html

Centre for Robotics Research, NMIT 8

 ROBOT-ANALOGOUS SYSTEM
❖ Controllers: This is the brain of the Robot, It receive the information, processes & Interprets the information & will
take suitable action.
• Example: Microprocessor, Microcontrollers

Analogy

Human Brain

Microcontrollers, Single Board Computers, Microprocessors

Centre for Robotics Research, NMIT 9

 ROBOT-ANALOGOUS SYSTEM
❖ Energy Source : These are the Source of Energy for the robot
• Example: Batteries, IC Engine etc

Analogy

Food Batteries, IC Engines etc

Adenosine triphosphate (ATP) is the source of energy for use and storage at the cellular level.

Centre for Robotics Research, NMIT 10

 ROBOT-ANALOGOUS SYSTEM
❖ Program: These are the set of Instructions given to the Robot to perform a specific task. Controllers will make use of
these Instructions to take decisions.
Analogy
Experience from
real world

Memory

Perception

Consciousness

Act

AI/ML

Centre for Robotics Research, NMIT 11

 Bio Mimicry of Locomotion
Types of Locomotion Challenges of Bio Mimicry of Locomotion
Biological System Artificial System
Cell division, in combination with Mechanical Structural Complexity. Each
Specialization, can readily produce a part must be individually fabricated-
millipede with several hundred legs Lack of economies of the scale
and several tens of thousands of
individually sensed cilia

Extreme miniaturization- very small Very difficult to meet the robustness of

size and weight, insects achieve a biological system using existing
level of robustness fabrication techniques & materials.

the biological energy storage system Man made structures with similar
and the muscular and hydraulic miniature scale have poor power to
activation systems – High Torque, weight ratios, poor conversion
Good Response time, and Conversion efficiency
efficiencies

Reference: Introduction to Autonomous Mobile Robots - Roland Siegwart and Illah R. Nourbakhsh A Bradford

Centre for Robotics Research, NMIT 12

 Bio Mimicry of Locomotion
Few Examples

Centre for Robotics Research, NMIT 13

 Bio Mimicry-Android Robot
" An Android Uncanny Valley By Masahiro Mori
Robot is a one
which has Human
Resemblance &
Behaviour"
Android Form Caricatured Form

v/s

Geminoid Nexi
Picture Courtesy: Picture Courtesy: MIT
www.Geminoid.jp Media Labs

Centre for Robotics Research, NMIT 14

 EXAMPLE-ANDROID ROBOT DESIGN
Machine-Like Maximalism

HRI
Human-Like Dynamics
Minimalism
Complex Design Simple Design
Longer Development Cycle Lower Development
Higher Development & Maintenance Cost & Maintenance Cost

Centre for Robotics Research, NMIT 15

 EXAMPLE-ANDROID ROBOT DESIGN
Machine-Like Maximalism

Design

Anthropomorphism

HRI
Human-Like Dynamics
Minimalism
Complex Design Simple Design
Longer Development Cycle Lower Development
Higher Development & Maintenance Cost & Maintenance Cost

Centre for Robotics Research, NMIT 16

 EXAMPLE-ANDROID ROBOT DESIGN
Machine-Like Maximalism

Anthropomorphism

Design

HRI
Human-Like Dynamics
Minimalism
Complex Design Simple Design
Longer Development Cycle Lower Development
Higher Development & Maintenance Cost & Maintenance Cost

How Minimalist Android Affects

Interaction Dynamics?

Centre for Robotics Research, NMIT 17

 EXAMPLE-ANDROID ROBOT DESIGN
ARIA-1st Gen

ADVANCED
ROBOT FOR
INTERACTIVE
APPLICATION

Centre for Robotics Research, NMIT 18

 EXAMPLE-ANDROID ROBOT DESIGN
Anatomical & Physiological Data Design
Inputs Pitch & Yaw Mechanism of Eyeball Mimicking Human Eye Movements
Inputs for Configuration &
Mechanics of Parametric Design
Vision not
considered since Structure Design
eyes in present U-Joint
context do not Kinematic Links
provide any Holder Hole to Fasten
vision to robot Mechanism Design
Establishment Pitch Spindle
Eye Movements of Functional
Requirements
Saccade Smooth Pursuit Convergence
Pitch Pitch
Discretization of Eye movements
Spindle
Pitch & Yaw Movement Yaw
Hole to
Design
Design Synthesis In terms of Degrees of Freedom (DoF) Connect Link
Synthesis
Phase
for 2 Eyes Universal Joint
2 DoF 3 DoF 4 DoF
(Yaw Spindle)

Convergence All 3 Eye All 3 Eye

Not Possible movements movements
Possible Possible

Eliminate Actuators Actuators

Design
 Development Analysis
Development And Test
Time
Time Phase
Design
Complexity Design
Simplicity

Eliminate
Accept
Design
Design
Indicates Decrease
 ✓ Indicates Increase
Indicates Decrease
Indicates Increase
Centre for Robotics Research, NMIT 19
 EXAMPLE-ANDROID ROBOT DESIGN
ARIA-1st Gen

Centre for Robotics Research, NMIT 20

 EXAMPLE-ANDROID ROBOT DESIGN
ARIA-2nd Gen Robot

CAD Model-Front 3D Printed Model Head with Teeth Head (Post Aesthetics) with
& loops Silicone Skin & Synthetic
Hairs

Centre for Robotics Research, NMIT 21

 EXAMPLE-ANDROID ROBOT DESIGN
Ameca-World’s Most Advanced Android Robot

Appropriate facial expressions in

this video are selected by GPT3 -
we also tried GPT4 , the
processing time with 4 was
longer and made Ameca appear
less responsive

https://www.engineeredarts.co.uk
/about-us/brochures/

Video Courtesy: Engineered Arts

Centre for Robotics Research, NMIT 22

 PRACTICAL ASPECTS OF ROBOT DESIGN
& CONTROL

Centre for Robotics Research, NMIT 23

PRACTICAL ASPECTS OF ROBOT DESIGN AND CONTROL
Popular Embedded
Practical Considerations Robot Architecture-Example
Boards/SBCs

• Processing Power
• Mechanical & Electrical
Characteristics of Motors
• Sensors
Arduino
• Power Consumption
• GPIO Pins
• Interface
• Form factor
• Communication Protocol
• GPU requirements
Raspberry Pi • Memory Requirements
• OS Support
• Signal Conditioning
System
• Support from OEM
• Cost
• Availability
Nvidia Jetson • Peripheral Support Ref: Rashad, Ramy. (2012). Dynamic trajectory tracking of a car-like robot. 10.13140/RG.2.2.17982.74567.

Centre for Robotics Research, NMIT 24

 DAWN OF ARTIFICIAL INTELLIGENCE

Centre for Robotics Research, NMIT 25

 DAWN OF ARTIFICIAL INTELLIGENCE

Courtesy: https://medium.com/ml-ai-study-group/ai-mind-map-a70dafcf5a48

Centre for Robotics Research, NMIT 26

 APPLICATION OF AI

APPLICATION OF AI IN FACTORIES OF FUTURE

Centre for Robotics Research, NMIT 27
 APPLICATION OF AI

APPLICATION OF AI IN FACTORIES OF FUTURE

Centre for Robotics Research, NMIT 28
 APPLICATION OF AI IN FACTORIES-EXAMPLES
French food manufacturer Danone Group uses machine learning to improve its demand forecast accuracy.
This has led to a:
•20% decrease in forecasting errors
•30% decrease in lost sales
•50% reduction in demand planners’ workload

Fanuc, a Japanese automation company, uses robotic workers to operate its factories round-the-clock. The robots can
produce essential components for CNCs and motors, operate all production floor machinery non-stop, and facilitate
continuous monitoring of all operations.

The BMW Group uses automated image recognition for quality checks, inspections, and to eliminate pseudo-defects
(deviations from target despite no actual faults). As a result, they’ve achieved high levels of precision in manufacturing.

Porsche use autonomous guided vehicles (AGVs) to automate significant portions of automotive manufacturing. The
AGVs take vehicle body parts from one processing station to the next, eliminating the need for human intervention and
making the facility resilient to disruptions like pandemics.

Centre for Robotics Research, NMIT 29

 APPLICATION OF AI IN FACTORIES-EXAMPLES

Courtesy: https://www.birlasoft.com/articles/17-use-cases-of-ai-in-manufacturing

Centre for Robotics Research, NMIT 30

 APPLICATION OF AI IN FACTORIES-EXAMPLES
Key Take Away
A McKinsey survey found that companies
embracing digital transformation in
manufacturing are leading the industry. They
adopted 4IR technologies, such as big data
analytics, AI, AR and VR, IoT, predictive analytics,
automation, and robotics, among others.

As a result, they recorded several benefits in

terms of manufacturing efficiency, productivity,
and costs. These include:

30-50% decrease in machine downtime

15-30% increase in labor productivity
10-30% improvement in throughput
10-20% reduction in quality-related costs

Courtesy: https://www.datasciencecentral.com/pros-and-cons-of-ai-in-manufacturing/

Courtesy: https://www.birlasoft.com/articles/17-use-cases-of-ai-in-manufacturing

Centre for Robotics Research, NMIT 31

 ANALYTICS+AI=SMART FACTORIES
Data Analytics

Image Courtesy: www.Principa.co.za

Centre for Robotics Research, NMIT 32

 ANALYTICS+AI=SMART FACTORIES
Data Analytics

Image Courtesy: www.Principa.co.za

Centre for Robotics Research, NMIT 33

 ANALYTICS+AI=SMART FACTORIES
Anamoly Detection-Example

Image Courtesy: www.Principa.co.za

Centre for Robotics Research, NMIT 34

 ANALYTICS+AI=SMART FACTORIES
Battery Management using AI in Factory

Centre for Robotics Research, NMIT 35

 MACHINE LEARNING
Traditional v/s Machine Learning Technique

Data (Input) Program Data (Input) Output

Traditional
Machine Learning
Programming
Technique
Technique

Output Program

Centre for Robotics Research, NMIT 36

 MACHINE LEARNING

The subfield of computer science that “gives computers the

ability tolearn without being explicitly programmed”.
(Arthur Samuel, 1959)

For Answering
Using Data
Questions

Training Predicting

Centre for Robotics Research, NMIT 37

 MACHINE LEARNING-TYPES

Supervised

Unsupervised

Reinforcement

38
Centre for Robotics Research, NMIT 38
 MACHINE LEARNING-TYPES

Learn through examples of which we know the

Supervised desired output (what we want to predict).

Is this a cat or a dog?

Are these emails spam or not?

Unsupervised
Predict the market value of houses, given the
square meters, number of rooms, neighborhood,
etc.
Reinforcement

39
Centre for Robotics Research, NMIT 39
 MACHINE LEARNING-TYPES

Supervised Classification

Output is a discrete
variable (e.g., cat/dog)

Unsupervised
Regression

Output is continuous
Reinforcement (e.g., price, temperature)

40
Centre for Robotics Research, NMIT 40
 MACHINE LEARNING-TYPES

Supervised Classification

Output is a discrete
variable (e.g., cat/dog)

41
Centre for Robotics Research, NMIT 41
 MACHINE LEARNING

What is a Dog?

Centre for Robotics Research, NMIT 42

 MACHINE LEARNING
Challenges associated with Machine Learning

Occlusion Diversity Deformation Lighting variations

Centre for Robotics Research, NMIT 43

 MACHINE LEARNING-TYPES

Supervised

There is no desired output. Learn somethingabout the

data. Latent relationships.

Unsupervised I have photos and want to put them in 20

groups.

I want to find anomalies in the credit card usage

patterns of my customers.
Reinforcement

Centre for Robotics Research, NMIT 44

 MACHINE LEARNING-TYPES

Supervised Useful for learning structure in the data(clustering),

hidden correlations, reduce dimensionality,etc.

Unsupervised

Reinforcement

Centre for Robotics Research, NMIT 45

 MACHINE LEARNING-TYPES

Supervised An agent interacts with an environment andwatches the

result of theinteraction.

Environment gives feedback via a positiveor

negative rewardsignal.
Unsupervised

Reinforcement

Centre for Robotics Research, NMIT 46

 MACHINE LEARNING-STEPS
Degree of Effort for ML
ML in a Nutshell

Knowledge Prerequisites

Courtesy: https://www.oak-tree.tech/blog/ml-models

Centre for Robotics Research, NMIT 47

 MACHINE LEARNING-STEPS
Detailed Steps in ML

Data Data Feature Algorithm Selection Making

Gathering Preprocessing Engineering & Training Predictions
Collect data from Clean data to have Making your data Selecting the right Evaluate the
various sources homogeneity more useful machine learning model model

Courtesy: https://www.oak-tree.tech/blog/ml-models

Centre for Robotics Research, NMIT 48

 MACHINE LEARNING-STEPS
Data Gathering

Few Repositories where dataset can be downloaded

• https://www.kaggle.com/datasets
• https://archive.ics.uci.edu/ml/index.php

https://labelyourdata.com/articles/what-is-dataset-in-machine-learning

Centre for Robotics Research, NMIT 49

 MACHINE LEARNING-STEPS
Data Pre-Processing

Is there anything wrong with thedata?

• Missing values

• Outliers

• Bad encoding (fortext)

• Wrongly-labeled examples

• Biased data

• Do I have many more samples of

one class than therest?

Need to fix/removedata?

Courtesy: https://research.aimultiple.com/data-preprocessing/

Centre for Robotics Research, NMIT 50

 MACHINE LEARNING-STEPS
Feature Engineering

What is a feature?

A feature is an individual measurable property of a

phenomenon being observed

Our inputs are represented by a set of features

Example
Toclassify spam email, features couldbe:

• Number of words that have beench4ng3d

likethis.
Buy ch34p drugs from the (2, 0, 3)
• Language of the email(0=English, 1=Spanish) ph4rm4cy now :) :) :) Feature
Engineering
• Number ofemojis

Courtesy: https://research.aimultiple.com/data-preprocessing/

Centre for Robotics Research, NMIT 51

 MACHINE LEARNING-STEPS
Algorithm Selection

Courtesy: @CEO_AISOMA

Centre for Robotics Research, NMIT 52

 MACHINE LEARNING-STEPS
Algorithm Selection & Training

Centre for Robotics Research, NMIT 53

 MACHINE LEARNING-STEPS
Algorithm Selection & Training

Courtesy: https://www.javatpoint.com/supervised-machine-learning

54
Centre for Robotics Research, NMIT 54
 MACHINE LEARNING-ALGORITHM vs MODEL

Algorithm Model
An “algorithm” in machine learning is a procedure that is run A “model” in machine learning is the output of a machine learning algorithm
on data to create a machine learning “model.” run on data.

Machine learning algorithms perform “pattern recognition.” A model represents what was learned by a machine learning algorithm.
Algorithms “learn” from data, or are “fit” on a dataset.
The model is the “thing” that is saved after running a machine learning
machine learning algorithms have a number of properties: algorithm on training data and represents the rules, numbers, and any other
Machine learning algorithms can be described using math and algorithm-specific data structures required to make predictions.
pseudocode.
Examples:
The efficiency of machine learning algorithms can be analyzed The linear regression algorithm results in a model comprised of a vector of
and described. coefficients with specific values.

Machine learning algorithms can be implemented with any The decision tree algorithm results in a model comprised of a tree of if-then
one of a range of modern programming languages statements with specific values.

The best analogy is to think of the machine learning model as a “program.”

The machine learning model “program” is comprised of both data and a
procedure for using the data to make a prediction

Centre for Robotics Research, NMIT 55

 MACHINE LEARNING-ALGORITHM vs MODEL

Algorithm Model
An “algorithm” in machine learning is a procedure that is run A “model” in machine learning is the output of a machine learning algorithm
on data to create a machine learning “model.” run on data.

Machine learning algorithms perform “pattern recognition.” A model represents what was learned by a machine learning algorithm.
Algorithms “learn” from data, or are “fit” on a dataset.
The model is the “thing” that is saved after running a machine learning
machine learning algorithms have a number of properties: algorithm on training data and represents the rules, numbers, and any other
Machine learning algorithms can be described using math and algorithm-specific data structures required to make predictions.
pseudocode.
Examples:
The efficiency of machine learning algorithms can be analyzed The linear regression algorithm results in a model comprised of a vector of
and described. coefficients with specific values.

Machine learning algorithms can be implemented with any The decision tree algorithm results in a model comprised of a tree of if-then
one of a range of modern programming languages statements with specific values.

The best analogy is to think of the machine learning model as a “program.”

The machine learning model “program” is comprised of both data and a
procedure for using the data to make a prediction

Centre for Robotics Research, NMIT 56

 MACHINE LEARNING-ALGORITHM vs MODEL
An algorithm is run on data to create a model.
Machine Learning => Machine Learning Model

We also understand that a model is comprised of both data and a procedure for how to use the data to make a prediction
on new data. You can think of the procedure as a prediction algorithm if you like.
Machine Learning Model == Model Data + Prediction Algorithm

This division is very helpful in understanding a wide range of algorithms.

For example, most algorithms have all of their work in the “algorithm” and the “prediction algorithm” does very little.

Typically, the algorithm is some sort of optimization procedure that minimizes error of the model (data + prediction
algorithm) on the training dataset.

Linear Regression:
Algorithm: Find set of coefficients that minimize error on training dataset
Model:
Model Data: Vector of coefficients
Prediction Algorithm: Multiple and sum coefficients with input row

Centre for Robotics Research, NMIT 57

 MACHINE LEARNING-ALGORITHM vs MODEL
Training-Fitting Data

Courtesy: https://www.kaggle.com/getting-started/166897 Courtesy: https://www.javatpoint.com/bias-and-variance-in-machine-learning

Centre for Robotics Research, NMIT 58

 MACHINE LEARNING vs DEEP LEARNING

Courtesy: https://data-flair.training/blogs/deep-learning-vs-machine-learning/
Courtesy: https://www.scs.org.sg/articles/machine-learning-vs-deep-learning

Centre for Robotics Research, NMIT 59

 MACHINE LEARNING vs DEEP LEARNING

100s of Different Dog Breeds:

Different Shape, Different Colors,
Different Sizes etc- Such type of
Data Set requires Deep Learning
for Training.

Centre for Robotics Research, NMIT 60

 ARTIFICIAL NEURAL NETWORKS (ANN)-OVERVIEW
The term "Artificial Neural Network" is derived from Biological neural networks that develop the structure of a human
brain. Similar to the human brain that has neurons interconnected to one another, artificial neural networks also have
neurons that are interconnected to one another in various layers of the networks. These neurons are known as nodes.

Natural Neurons ANN Model ANN Architecture

Biological Neural Network Artificial Neural Network

Dendrites Inputs
Cell nucleus Nodes
Synapse Weights
Axon Output Courtesy: https://www.javatpoint.com/artificial-neural-network

Centre for Robotics Research, NMIT 61

 EXPERT SYSTEM
Expert System : The other name of expert systems is knowledge-based systems. They are used for the real-world problems
like expert quality advice, diagnosis and recommendations. Basically, it is a type of computer program that is used to simulate
the judgement and behavior of humans or an organization that has an expert knowledge and experience about the particular
field. Building of an expert system requires a human expert that extract the required knowledge.

Centre for Robotics Research, NMIT 62

 APPLICATION OF AI IN ROBOTS-CASE STUDY

Courtesy: https://www.youtube.com/watch?v=pQuUW3Jp8ic&list=PL_r4rS7sBXUJUBmoPra9vMKZ6clpg_tdO

Centre for Robotics Research, NMIT 63

 ABOUT US -CENTRE FOR ROBOTICS RESEARCH

Centre for Robotics Research is

one of the Centres of Excellence
at NMIT. It Comprises Group of
Researchers and Academicians
working on Frontier Areas of
Robotics & Allied Disciplines.

THRUST AREAS OF RESEARCH

• Design & Development of Robots & Mechanisms

• Embedded Systems/Control Engineering
• Artificial Intelligence & Machine Learning
• Mechatronics
• Industrial Automation/Robot Process Automation (RPA)
• Signal Processing
• Computer Vision/Machine Vision
• Unmanned Ground/Aerial Vehicles/ Drones
• Social Robotics/Human Robot Interaction
• STEM/STEAM Education using Robotics Kit

Centre for Robotics Research, NMIT 64

 ABOUT US -CENTRE FOR ROBOTICS RESEARCH-TRAINING

Park Uneven
Bell Temple Terrain

See Saw
Flower

Waste Segregation
Black

Swing

White
Shopping Mall

Centre for Robotics Research, NMIT 65

 ABOUT US -CENTRE FOR ROBOTICS RESEARCH-TRAINING

Centre for Robotics Research, NMIT 66

 ABOUT US -CENTRE FOR ROBOTICS RESEARCH-TRAINING

Centre for Robotics Research, NMIT 67

 THANK YOU
ANY QUESTIONS?

Centre for Robotics Research, NMIT 68