Computer science Notes 0478 ​

​
Chapter 6 - Automated System, Robotic & Artificial Intelligence

Automated Systems

An automated system is a combination of software and hardware

designed and programmed to work automatically without the need for
human intervention.

Unlike simple hardware, automated systems don't require manual

adjustments. They operate independently once programmed.
There are three important hardware components that allow automated
systems to work:

●​ Sensor: Known as the automated input, it takes readings from

the surrounding environment and sends data to the
microprocessor.
●​ Microprocessor: Processes input from sensors and sends
instructions to other components like actuators. It acts as a
middleman.
●​ Actuator: Turns electrical signals from the microprocessor into
output, such as movement or activation of devices.

If the sensor captures analog data, such as temperature or density, it

must be converted into a digital format (0s and 1s) before the
microprocessor can process it.

Applications of Automated Systems

Here are some applications of automated systems, although the list
isn't exhaustive.
Transport (Autonomous Cars) -
●​ Sensors: Act as the "eyes and ears" of the car, detecting the
surroundings.
●​ Microprocessor: Processes data from sensors to determine the
next actions.
●​ Actuators: Perform actions based on the sensor data, such as
steering or accelerating.

Advantages Disadvantages
Expensive due to computer
Safer due to precise calculations
involvement
More efficient operation, leading to a Risk of hacking into the
better environment vehicle's control system

Agriculture (Automated Harvesting) -

●​ Sensors: Detect optimal harvesting times based on factors like

weather conditions.
●​ Microprocessor: Determines where to harvest based on sensor
data.
●​ Actuators: Carry out the extraction of goods from the ground.

Advantages Disadvantages
More accurate delivery of chemicals, Fewer job opportunities
ensuring successful farming due to automation

Weather Forecasting -

●​ Sensors: Check for dangerous weather conditions, measuring

humidity and wind speed.
 ●​ Microprocessor: Issues relevant warnings based on gathered
data.
●​ Actuators: Activate emergency systems.

Advantages Disadvantages
Constant information about overhead Can be inaccurate,
weather conditions is fed to air traffic especially with snow
controllers forecasts

Gaming -

●​ Sensors: Built into game environments to track interactions with

human controllers.
●​ Microprocessor: Determines game outcomes and decisions.
●​ Actuators: Activated in controllers to provide haptic feedback.

Advantages Disadvantages
Deeper gaming challenges with Decreased social
computer-generated opponents behaving interaction with other
more like humans humans

Lighting -

●​ Sensors: Detect the amount of light in the surrounding area.

●​ Microprocessor: Runs routines based on human requirements
(programmed rules).
●​ Actuators: Activate home equipment such as turning lights on or
off.
Advantages Disadvantages
System can react to environmental Smart bulbs and systems can be
factors automatically expensive to maintain

Science (Automated Laboratory Experiments) -

●​ Sensors: Detect changes in environmental conditions like

humidity.
●​ Microprocessor: Calculates alterations needed for the
experiment settings.
●​ Actuators: Perform actions like turning on heaters, coolers, and
water supplies.

Advantages Disadvantages
More experiments can be repeated, and Disadvantage not
data readings can take place faster than mentioned in lecture
manual tests transcript

Robotics
Robotics involves the design, construction, and operation of robots to
enhance our working and personal lives.

Characteristics of a Robot :
To be considered a robot, a technology must possess the following
three characteristics:
1.​Electrical Components: Includes sensors, microprocessors, and
actuators. Robots are a subset of automated systems.
 2.​Mechanical Structure: Machinery with movable parts like wheels,
gears, and pistons to carry out functions such as turning,
twisting, moving, gripping, or lifting.
3.​Programmable: Robots can be programmed with code to dictate
how they handle input data, enabling them to perform specific
tasks.

Independent vs. Dependent Robots

●​ Independent Robots: Operate autonomously without direct
human control (e.g., autonomous vehicles).
●​ Dependent Robots: Require a human to interface directly with
and control the robot.

Applications of Robotics
Robots are used across various industries such as :

●​ Car Industry: Robots assemble cars and work in hazardous

conditions.
●​ Transport: Autonomous cars and Mars rovers enhance safety
and environmental impact.
●​ Agriculture: Robots perform repetitive tasks like harvesting,
allowing farmers to focus on improving production.
●​ Medicine: Robots are used in surgeries and for research and
development, such as monitoring patient data and taking blood
samples.
●​ Domestic: Robots like vacuum cleaners and lawn mowers
automate household tasks.
●​ Entertainment: Robots are used in the film industry to capture
shots that humans cannot.
Advantages and Disadvantages of Robotics:

Aspect Advantages Disadvantages

Can work in hazardous May lead to higher
conditions, operate unemployment among manual
General
continuously, and perform laborers and risk de-skilling
tasks with high accuracy. when robots take over tasks.
Replaces slow, repetitive
Agricult tasks, allowing farmers to
Fewer job opportunities.
ure concentrate on improving
production.
Transpo Safer and better for the
Can be expensive to set up.
rt environment.
Surgeries take less time,
Medicin improving patient recovery
Expensive.
e chances; robots can spot
veins better than humans.
Quickly learn the
Domesti environment and can be Can encourage a sedentary
c programmed to work at lifestyle.
convenient times.
Fewer job opportunities for
Entertai Films can be presented in photographers; increasing
nment ways not normally possible. replacement of humans by
machinery.

Impact on Human Tasks

One advantage of automated systems and robotics is that they free up

time for humans to focus on creative and innovative aspects of
business. By delegating repetitive tasks to robots, humans can
concentrate on new inventions and innovative ideas.

Defining Artificial Intelligence

The definition of AI is:

A branch of computer science dealing with the simulation of intelligent
human behavior by a computer.
Essentially, AI aims to instill human intelligence into a computer.
Intelligence is:
The ability to think, to learn from experience, to solve problems, and to
adapt to new situations.
AI aims to have computers learn all these things. We want the
computer to have our thought process inside its memory by instilling
our thought process into a machine.

Reasoning in AI

Here are three methods AI uses to reason:

1.​Drawing conclusions based on given data: For example,
predicting house prices based on square footage using historical
data. This can be done using linear regression, which draws a
line of best fit based on existing data to predict outcomes.
2.​y=mx+b
3.​y=mx+b
4.​Learning rules: Teaching machines to follow certain rules, like
detecting if someone has COVID-19 by asking questions such as
"Does he cough?" or "Does he have a fever?".
5.​Pattern recognition: Quickly discerning patterns and making
predictions by adapting to new data, such as detecting unusual
bank transactions based on historical data.
Branches of AI
There are two main branches of AI:
●​ Expert Systems
●​ Machine Learning

Expert Systems

An expert system is a computer system that mimics the

decision-making ability of a human.

It simulates the judgment and behavior of a human or organization

with expert knowledge and experience. An example is a doctor who
diagnoses diseases based on symptoms.

Structure of an Expert System

Every expert system should have:

●​ User Interface: Allows humans to interact with the system.

●​ Knowledge Base: Stores knowledge about an area of expertise
from expert sources.
●​ Rule Base: Rules used by the inference engine to draw
conclusions.
●​ Inference Engine: Makes decisions based on the knowledge and
rule base to produce possible solutions.
●​ Explanation System: Suggests actions for the user and provides
the accuracy of predictions.
Components of an Expert System :
Compone Example (Medical
nt Description Diagnosis)
Stores knowledge about an
Knowledg area of expertise obtained A library of symptoms
e Base from a number of expert related to diseases.
sources.
Uses certain rules used by If multiple symptoms are
Rule
the inference engine to draw present, suggest possible
Base
a conclusion diseases.
Suggest possible diseases
Explanati Suggests certain actions for
and action steps; provide
on the user to take and provides
the probability of
System the accuracy of its prediction
COVID-19.

Expert System Examples -

Explanation
Application Knowledge Base Rule Base System
Bank Loan If credit score is Explain reasons
Financial data,
Approval above a threshold, for approval or
credit history
System approve loan denial
If certain
Suggest
Mineral geological
Geological data, locations for
Prospectin formations are
mineral locations further
g present, predict
exploration
mineral deposits
Mammography If patterns in
Predict
Diagnosis images with mammogram
probability of
of Breast cancer and match cancer
cancer based
Cancer normal indicators, predict
on mammogram
mammography probability
Advantages of Expert Systems :
●​ High level of expertise
●​ High accuracy
●​ Consistent results (no emotions involved)

Machine Learning: An Overview

Machine learning is a category of artificial intelligence that differs from

expert systems. Instead of relying on predefined rules, machine
learning algorithms use data points to create models that can predict
outcomes.
Machine learning is the science of training computers with sample
data so that they can make predictions about new, unseen data
without being specifically programmed for those new data instances.

Key Aspects of Machine Learning:

●​ Algorithms are trained and learn from past experiences and

examples.
●​ No need for explicit rule-based programming (like if/then
statements).
●​ Ability to make predictions about new, unseen data.

Spam Email Classification: A Practical Example

Email services like Google use machine learning to classify emails as

spam. Instead of manually defining rules, machine learning algorithms
collect data about emails.
 1.​Data Collection: Gather data about emails.

2.​Cleaning Process: Remove Stop Words: Eliminate frequently

appearing words like "the," "and," and "a" that don't help
determine if an email is spam.

3.​Model Building: Create a machine learning model using the data.

4.​Training Data Set: Train the model using past emails known to
be spam. This involves feeding the computer millions of spam
emails to learn how to categorize them.

5.​Evaluation and Fine-Tuning: Evaluate the model, fine-tune it, and

test it live.

Auto-Tagging in Photos
Another application is auto-tagging in photo libraries (like Apple
Photos), where the system recognizes faces and suggests names.
The system identifies similar faces in your photos.

AlphaGo: Mastering the Game of Go

AlphaGo is a machine learning algorithm designed to play the game
Go. The algorithm is fed tons of Go matches. It then uses this
knowledge to determine the best move in a given scenario. AlphaGo
even beat world champion Lee Sedol in a match.

Expert Systems vs. Machine Learning

Feature Expert Systems Machine Learning

Approach Rule-based Statistical modeling of data
 Functions using Uses machine learning models
Method
predefined rules (linear regression, etc.)
Decision Relies on "if-then" Projects input into model spaces
Making statements to predict outcomes

Machine Learning Models

●​ Linear Regression
●​ Regression
●​ Decision Tree

These models are used by machine learning engineers.