Chapter 6: Automated & Emerging Technologies

#	Topic
6.1	Automated Systems
6.2	Robotics
6.3	Artificial Intelligence
6.4	Expert Systems
6.5	Machine Learning
—	Past Paper Questions and Model Answers
—	Flash Cards
—	Glossary
—	Exam Command Words
—	The SCAR Formula

6.1 Automated Systems

What is it?
An automated system is a machine that works on its own — no human
needs to press any buttons or make any decisions.
It uses sensors to collect information, a microprocessor to
think about it, and actuators to take action.

Think of it like this: When your phone screen gets brighter when
you step outside, you didn't do anything. A tiny sensor spotted the
bright light. The processor thought "it's bright out there." The screen
got brighter. All in less than a second — completely automatic.

How It Works — The Control Loop

The system keeps doing the same thing over and over. This never-ending
cycle is called a feedback loop. Here is how it works step by step:

Step	Part	Simple Explanation	Real Example
1	Sensor	Notices something in the real world	Camera on Tesla detects a car 15m ahead
2	ADC	Translates the signal into numbers the processor can read	Converts the distance measurement into binary
3	Microprocessor	Checks if that number is within the safe range	Is 15m close enough to brake?
4	Decision	Fine = do nothing. Not fine = tell the actuator	Yes, too close — brake now
5	Actuator	Does something physical in the real world	Brakes squeeze — car slows down
6	Feedback	Starts again immediately	Now how far away is the car?

What Is the ADC and Why Do We Need It?

Here is the key thing to understand.

Sensors record the real world. The real world is analogue — things
change smoothly and gradually. Temperature doesn't jump from 20°C to
21°C in one snap. It drifts slowly through every tiny value in between.

But computers only understand digital — 0s and 1s. Nothing in between.

The ADC (Analogue-to-Digital Converter) is the translator.
It turns the smooth real-world signal into the digital steps a
computer can actually read.

KEY EXAM POINT
If a sensor produces analogue data (and most do), you MUST
mention the ADC in your exam answer. Without the ADC,
the microprocessor cannot read the sensor data.
It only understands binary. Always say the process is continuous.

Common Sensors — What They Detect

Sensor	What It Detects	Where You See It
Temperature	How hot or cold something is	Google Nest thermostat, greenhouse heating
Light (LDR)	How bright the light is	Phone screen auto-brightness
Pressure	Force being applied	Supermarket automatic sliding doors
Motion (PIR)	Movement nearby	Ring doorbell, security lights
Humidity	How much moisture is in the air	Greenhouse climate control
Moisture	How wet the soil is	Automatic garden sprinklers
Gas / Chemical	Harmful gas levels	Carbon monoxide detector at home
Sound (Acoustic)	Sound or vibrations	Security system detecting broken glass
Proximity / Ultrasonic	Distance to nearby objects	Self-parking car
pH	How acidic a liquid is	Water treatment plant
Accelerometer	Sudden movement or impact	Car airbag — fires when crash detected
Pulse	Heart rate	Apple Watch health tracking

Good Things and Bad Things About Automated Systems

Good Things (Advantages)	Bad Things (Disadvantages)
Faster and more accurate than humans	Costs a lot of money to set up
Spots problems and acts straight away (real-time)	Can be hacked by criminals
Cheaper to run long-term than paying workers	People stop learning skills because machines do everything
Works all day and all night — never tires	Cannot handle unusual or unexpected situations
Same quality output every single time	If one part breaks, the whole system can stop
Safe to use in dangerous places humans cannot go	Needs experts to fix when things go wrong

Where Automated Systems Are Used

Industry

Amazon's robots pack and sort millions of orders every day

Robotic arms in Tesla's factory weld car parts with perfect precision

Cameras on conveyor belts spot damaged products and remove them automatically

Transport

Tesla Autopilot: sensors watch the road every millisecond and control the car

Airplane autopilot: keeps the plane at the right height and speed automatically

Self-parking cars: sensors scan the space, then the car steers itself in

Agriculture

Soil sensors check if plants need water and open valves automatically

Drones scan crops from above and spray only the parts that need treatment

Autonomous tractors drive themselves across fields using GPS

Weather

Weather stations measure temperature and wind, sending data to forecasters

Flood sensors trigger flood gates automatically when water levels get too high

Gaming

VR headsets track your head movement and update the screen instantly

Motion controllers turn your real arm movements into in-game actions

Lighting

Street lights switch on at dusk when the LDR detects it getting dark

Security lights flick on when the PIR sensor detects someone moving nearby

Science

Nuclear power plants shut themselves down automatically if radiation spikes

Hospital drip machines adjust the dose based on the patient's pulse sensor

6.2 Robotics

What is it?
Robotics is the subject that covers how to design, build, and control robots —
physical machines that can move and do things in the real world.

Think of it like this: A Roomba vacuums your floor by itself. Amazon
robots carry shelves across warehouse floors without anyone driving them.
A robotic arm in a factory welds the same car part thousands of times a day
without getting tired or making mistakes. These are all robots.

Three Things Every Robot Must Have

Part	What It Means	Real Example
Mechanical Structure	The physical body — it can move and touch things	Boston Dynamics Spot's four legs
Electrical Components	The brain and senses — sensors, processor, motors	Cameras, computer chip, hydraulic legs
Programmable	It follows stored instructions — and those can be changed	Amazon robot programmed with a floor map

IMPORTANT EXAM NOTE
Robots do NOT automatically have AI.
Most factory robots just repeat the same movement over and over.
They follow pre-written instructions — they do NOT think or learn.
AI is only there if someone has specifically added it.

Two Types of Robot

Type	What It Means	Does It Replace Humans?	Example
Independent	Works completely on its own — no human needed	Yes, completely	Amazon warehouse robot, Roomba
Dependent	A human controls it using a screen or joystick	No — the human is still in charge	da Vinci surgical robot

Where Robots Are Used

Good Things and Bad Things About Robots

Good Things (Advantages)	Bad Things (Disadvantages)
Works 24/7 with no breaks or holidays	Cannot handle surprises or unusual situations
Can work in dangerous places humans cannot	Puts people out of work — unemployment rises
More accurate and consistent than any human	People lose skills because robots do the work
Cheaper in the long run than paying wages	Very expensive to buy and set up
No mistakes caused by being tired or distracted	Can be hacked and taken control of

6.3 Artificial Intelligence

What is it?
AI is about making computers behave in a way that seems intelligent —
like reasoning, making decisions, learning from experience, and spotting patterns.

Think of it like this: TikTok's For You Page feels like it reads your mind
after just one day. Gmail spots a scam email before you even open it.
Spotify makes a playlist that sounds like your best friend built it for you.
ChatGPT answers questions it was never directly taught.
That is all AI — and you use it every single day.

EXAM SCOPE NOTE
For IGCSE, you only need to know AI through two things:
Expert Systems and Machine Learning.
That is it. Nothing else is required.

Five Characteristics of AI

These are the things that make AI different from normal programs:

Characteristic	Simple Explanation	Real Example
Simulates human behaviour	Acts like a thinking human	ChatGPT writing an essay on a topic it wasn't directly taught
Uses data and rules	Learns from millions of examples and follows logical rules	Gmail spam filter trained on millions of real spam emails
Can reason	Joins facts together to reach a conclusion	Bank system: unusual time + foreign country = block the card
Learns and adapts	Gets better by itself over time	Netflix recommendations improve the more you watch
Analyses patterns	Spots repeated trends in huge amounts of data	TikTok noticing you always stop scrolling for cooking videos

6.4 Expert Systems

What is it?
An expert system is a computer program that works like a human expert in one
specific subject. You ask it questions and it gives you professional advice —
just like a doctor, lawyer, or engineer would.

Think of it like this: When your bank blocks your card at 3am because someone
tried to use it abroad, no human at the bank did that. A rule fired automatically:
IF the time is unusual AND the location is foreign AND the amount is above normal
THEN block the card.
Done. No human. Instant.
That is an expert system.

The Four Parts of an Expert System

How the Inference Engine Works — Step by Step

The Four Parts — Easy to Remember

Part	What It Does	Simple Analogy
Knowledge Base	Stores all the facts and information	A giant textbook with every fact ever written about the subject
Rule Base	Stores all the IF-THEN decision rules	A list of instructions like "if it quacks and swims, it's a duck"
Inference Engine	Reads the facts, applies the rules, reaches a conclusion	The doctor reading your test results and making a diagnosis
Interface	Where you ask questions and get your answers	The website form, app screen, or chatbot you type into

EXAM TIP
Learn exactly these four parts. Past papers have directly asked you
to fill in the blanks with these exact four words.
Knowledge Base. Rule Base. Inference Engine. Interface.

Where Expert Systems Are Used

Subject	What It Does	Real Example
Medicine	Diagnoses illnesses from symptoms	MYCIN — diagnosed bacterial infections in the 1970s
Banking	Spots and blocks fraudulent transactions	Revolut, Visa blocking dodgy card payments instantly
Law	Predicts court outcomes from past cases	Legal AI tools used in UK and US
Agriculture	Identifies crop diseases from photos	Plant disease apps used by farmers across India and Africa
Tech Support	Diagnoses your problem before a human speaks to you	Apple Support bot, broadband troubleshooting tool
Education	Makes your practice questions easier or harder based on your scores	Khan Academy's personalised learning

Good Things and Bad Things

Good Things (Advantages)	Bad Things (Disadvantages)
Works like a real expert — accurate and consistent	Costs a huge amount to build and keep updated
Available 24 hours a day, 7 days a week	Only knows what it was taught — nothing outside that
Answers questions faster than any human expert	Feels cold and robotic — no care or empathy
Never retires — expert knowledge is saved forever	Cannot use common sense — only applies its rules
Same quality answer every single time	Has to be regularly updated as new knowledge is discovered

6.5 Machine Learning

What is it?
Machine Learning is a type of AI where the computer teaches itself by
looking at lots of examples. Nobody writes the rules for it.
It figures the rules out on its own from the data.

Think of it like this: Spotify's Discover Weekly playlist.
Nobody at Spotify manually picked those 30 songs for you.
Spotify's system watched 600 million users — what they played, skipped,
rewound, and saved — spotted patterns, and built a playlist just for you.
A new one, every Monday, automatically.
That is machine learning.

Normal Programming vs Machine Learning

Where Machine Learning Is Used

What It Does	Where You See It
Recommends what to watch or listen to next	Netflix, Spotify, YouTube, TikTok
Spots and blocks spam emails	Gmail — stops billions of spam emails every single day
Recognises your face to unlock your phone	iPhone Face ID
Detects fraud the moment a dodgy payment is made	Revolut, Visa, PayPal
Understands spoken words and commands	Siri, Alexa, Google Assistant
Finds cancer in medical scans before a doctor might	Google DeepMind working with the NHS
Calculates the fastest route based on live traffic	Google Maps, Waze

AI vs Machine Learning — What Is the Difference?

	Artificial Intelligence	Machine Learning
What it is	Making computers act intelligently	A specific type of AI that learns from data
Who writes the rules?	Human experts write the rules	The system finds its own rules from examples
Does it change over time?	Not always — expert systems stay fixed	Yes — adapts automatically with new data
Example	Bank fraud expert system	TikTok For You Page, Netflix recommendations
Relationship	The bigger subject	A smaller part that lives inside AI

Past Paper Questions and Model Answers

Q1 — Greenhouse Temperature [8 marks]

Describe how sensors and a microprocessor keep a greenhouse between 25°C and 30°C.

Model Answer

A temperature sensor reads the temperature continuously

The data is sent to the microprocessor

Because the data is analogue, it goes through the ADC to be converted to digital

The microprocessor compares the reading to its stored safe range: 25°C to 30°C

Temperature below 25°C → signal sent to switch on the heater / close the vents

Temperature above 30°C → signal sent to switch off the heater / open the vents

Temperature within range → nothing happens

This whole process repeats over and over — it is the feedback loop

SCAR: Sensor. Compare. Actuator. Repeat.
Always mention ADC. Always say process is continuous.
Always state what happens when the value is fine (no action).

Q2 — Underground Mine [4 marks]

Describe how an automated system monitors temperature and oxygen in a mine.

Model Answer

Sensors for temperature and oxygen keep taking readings

The ADC converts the analogue data to digital so the processor can read it

Temperature drops too low → processor tells the heater to turn on

Oxygen drops too low → processor opens the oxygen valve

If levels cannot be restored, an alarm is triggered

The loop repeats continuously — always watching

Q3 — Garden Sprinkler [6 marks]

Describe how a microprocessor controls an automatic garden sprinkler.

Model Answer

A soil moisture sensor detects how dry the soil is

Data goes to the microprocessor — through the ADC first as it is analogue

Microprocessor compares the reading to a stored safe moisture level

Soil too dry → signal sent to the actuator (water valve)

Valve opens — sprinkler turns on and waters the garden

Soil moisture is fine → nothing happens

The process repeats continuously in a feedback loop

Q4 — Characteristics of AI [3 marks]

Describe the characteristics of AI. (Any 3 earn full marks)

Model Answer

Simulates human behaviour — the machine acts and decides like a human would

Uses data and rules — trained on huge datasets with logical rules guiding decisions

Can reason — joins facts together using logic to reach a conclusion

Learns and adapts — changes its own processes over time without reprogramming

Analyses patterns — spots trends in data that humans would never notice

Q5 — AI in a Robot [4 marks]

Explain how a robot uses AI.

Model Answer

The robot uses AI to make decisions from sensor data — not just follow fixed steps

It uses rules learned from training data to decide what action to take

Over time it learns from what happens and gets better at its job

AI means it can cope with new situations it was never directly taught about

Q6 — Two Parts of an Automated System [2 marks]

State two main components of an automated system.

Model Answer

Sensor — collects data from the physical environment

Microprocessor — processes the data and decides what to do

(Also fine to write: Actuator / ADC)

Q7 — Robots in a Car Factory [3 marks]

Give THREE advantages of using robots on a car assembly line.

Model Answer

Robots work 24/7 without any breaks — more cars built consistently

Robots are more precise — fewer mistakes in welding and painting

Robots can work in dangerous conditions — fumes and heat that would harm humans

Q8 — Expert System Fill in the Blanks [3 marks]

(Real W23 Past Paper Question)

A car repair garage uses an expert system.
Use these words to complete the sentences:
inference engine / interface / knowledge base / rule base

Model Answer

The mechanic uses the interface to type in the symptoms and get a recommendation

The knowledge base stores all the facts about car faults and what causes them

The inference engine looks at the facts, applies the rules, and works out the diagnosis

Q9 — Machine Learning [2 marks]

(Real W23 Past Paper Question)

The expert system has machine learning capabilities.
What does this mean?

Model Answer

The system can automatically update and improve itself based on new cases it sees

It does this without a human reprogramming it — it teaches itself from experience

Q10 — AI vs Machine Learning [3 marks]

State the difference between AI and Machine Learning.

Model Answer

AI means making machines act intelligently; ML is a type of AI where
the machine learns from data automatically

In AI, human experts write the rules; in ML the machine
figures out its own rules from examples

AI systems may stay fixed after being built; ML systems
keep adapting as new data arrives

Q11 — Expert System Components [4 marks]

Name and describe TWO components of an expert system.

Model Answer

Knowledge Base — stores all the facts and information about the subject
(e.g. every known symptom and disease)

Inference Engine — the part that reads the facts, applies the IF-THEN
rules, and figures out the answer or recommendation for the user

Q12 — Self-Parking Car [4 marks]

A self-parking car uses ultrasonic sensors.
Describe how it works.

Model Answer

Ultrasonic sensors send out sound pulses and measure how far away obstacles are

Data goes to the microprocessor — through the ADC to convert it to digital

Processor compares the distance to a stored safe minimum value

Too close → signals the actuators (steering and brakes) to adjust position

The loop repeats continuously until the car is safely parked

Q13 — ATMs vs Human Bank Tellers [5 marks]

One advantage and one disadvantage of an ATM compared to a human cashier.

Model Answer

Advantage

ATMs work 24 hours a day, 7 days a week — you can get cash at any time,
not just during bank opening hours

They handle simple transactions faster than a human cashier

Disadvantage

ATMs can break down or get hacked — leaving you with no way to get your money

Using ATMs instead of people means bank tellers lose their jobs and customers
forget how to do basic in-branch banking

Q14 — Specimen Paper 1B Q10(a) — Characteristics of AI [3 marks]

An expert system is an example of artificial intelligence.
Describe what is meant by artificial intelligence.

Model Answer (any 3 from the list below)

AI simulates intelligent human behaviour in machines

AI systems have a collection of data and rules that guide their decisions

AI has the ability to reason — it uses logic to reach conclusions from facts

AI has the ability to learn and adapt — it updates itself based on experience

AI can analyse patterns — it spots trends in large amounts of data

Exam Tip: This question is worth 3 marks.
Write 3 separate, clear points. One short sentence each is enough.
Use the exact words from the characteristics list — simulates, reason,
learn and adapt, analyse patterns.

Q15 — Specimen Paper 1B Q10(b) — Expert System Components [3 marks]

One component of an expert system is the knowledge base.
Identify the three other components present in an expert system.

Model Answer

Rule Base

Inference Engine

Interface

Exam Tip: These exact four words appear in fill-in-the-blank questions.
Knowledge Base. Rule Base. Inference Engine. Interface.
Learn all four. Spell them correctly.

Q16 — Specimen Paper 1B Q10(c) — Machine Learning [1 mark]

An expert system can make use of machine learning.
State what is meant by machine learning.

Model Answer

Machine learning is when a program has the ability to
automatically adapt its own processes and/or data
based on experience — without being manually reprogrammed

Exam Tip: This is only 1 mark. One clear sentence is enough.
The key phrase the mark scheme wants is:
"automatically adapt its own processes and/or data"

Q17 — June 2024 Paper 12 Q10(a) — Garage Expert System [3 marks]

A garage uses an expert system to help diagnose problems with cars.
The expert system is an example of artificial intelligence.
Describe one characteristic of AI.

Model Answer (any 1 of these earns full marks)

AI has the ability to reason — it uses rules and logic to reach conclusions,
for example concluding that a car has a flat battery from the symptoms given

AI has the ability to learn and adapt — it updates its own decision-making
based on new cases it has seen, without being reprogrammed

AI can simulate human behaviour — it makes decisions the way a human
expert mechanic would, based on knowledge and rules

Q18 — June 2024 Paper 12 Q10(b) — How the Garage Expert System Works [6 marks]

The expert system is an example of artificial intelligence.
Explain how the expert system operates.

Model Answer

The mechanic types their question or the car's symptoms into the interface

The inference engine takes this input and searches the knowledge base
for relevant facts about car faults and their causes

The inference engine then applies the IF-THEN rules stored in the
rule base to those facts

For example: IF the engine makes no sound AND the lights are dim
THEN the battery is likely flat

The inference engine uses this process to reach a conclusion or recommendation

The diagnosis is sent back to the mechanic through the interface

Exam Tip: For a "how does it operate" question on expert systems,
always name all four parts and say what each one does.
Show the flow: Interface → Inference Engine → Knowledge Base + Rule Base → Interface.

Q19 — Oct/Nov 2023 Paper — Describe How an Expert System Operates [6 marks]

AI system is an expert system. Explain how an expert system operates.

Model Answer

The user interacts with the expert system through the interface —
they type in their question or situation

The inference engine receives the input and examines the knowledge base
to find facts that are relevant to the query

It then applies the IF-THEN rules from the rule base to those facts

The inference engine works through the rules until it reaches a conclusion

For example: IF the patient has a fever AND a cough AND breathlessness
THEN consider pneumonia

The conclusion or recommendation is returned to the user through the interface

Q20 — Machine Learning Improving a Robot [3 marks]

A farmer uses a robot to plant seeds in a field.
The robot is adapted to have machine learning capabilities.
Explain how this will improve the robot.

Model Answer

The robot can learn from its past planting sessions — for example
adjusting depth or spacing based on which seeds grew best

It can automatically adapt its processes without a farmer or programmer
having to manually update its instructions

Over time the robot becomes more accurate and more efficient — improving
its own performance based on patterns it has identified in real data

Exam Tip: The word is "explain" — so do not just say what ML is.
Say HOW it makes the robot better. Use the word "because" or "so that"
to force yourself to give a reason.

Q21 — Why Does an Expert System Need a Knowledge Base? [2 marks]

An expert system is an example of artificial intelligence.
Explain why an expert system needs a knowledge base.

Model Answer

The knowledge base stores all the facts and information about the subject
that the expert system needs to make decisions
(e.g. every known car fault, every known disease and symptom)

Without the knowledge base, the inference engine would have no facts
to work with — it cannot apply its rules to nothing, so it cannot
reach any conclusion or give any advice

Q22 — State the Difference Between an Expert System and Machine Learning [2 marks]

State one difference between an expert system and machine learning.

Model Answer (any 1 of these)

An expert system uses rules written by a human expert;
machine learning writes its own rules by finding patterns in data

An expert system's knowledge stays fixed unless a human updates it;
machine learning automatically adapts as new data arrives

An expert system cannot learn from new cases on its own;
machine learning improves itself over time without reprogramming

Q23 — Advantages and Disadvantages of Expert Systems [4 marks]

Give two advantages and two disadvantages of using an expert system.

Model Answer

Advantages (any 2)

Available 24/7 — never off sick, never on holiday

Gives consistent and unbiased results every single time

Can store and apply more knowledge than any single human expert

Responds faster than a human expert would

Disadvantages (any 2)

Very expensive to build and maintain

Can only use knowledge already stored in it — cannot handle
questions outside its domain

Gives cold, robotic answers — no empathy or human judgement

Needs regular updates or the knowledge becomes outdated

Q24 — Supervised vs Unsupervised Machine Learning [2 marks]

State the difference between supervised and unsupervised machine learning.

Model Answer

Supervised machine learning trains on data that has already been
labelled — for example emails pre-labelled as spam or not spam.
The system learns from the correct answers.

Unsupervised machine learning trains on data that has no labels —
the system finds its own patterns and groups in the data
without being told what to look for

Note: This point comes from the full syllabus content at
the bottom of the official notes. It appears in some
higher-level papers. Learn it as a bonus point.

Q25 — ATM Proximity Sensor [1 mark]

(Feb/Mar 2025 — 0478/12/F/M/25 Q4a-i)

A sensor is used to detect when a person stands within 1 metre of the ATM.
Identify an appropriate sensor for the ATM to detect a person.

Model Answer

PIR sensor (Passive Infrared sensor)

(Also accept: proximity sensor / ultrasonic sensor)

Q26 — Role of Microprocessor in ATM Detection [3 marks]

(Feb/Mar 2025 — 0478/12/F/M/25 Q4a-ii)

Describe the role of the microprocessor when the ATM detects a person.

Model Answer

The microprocessor receives the signal from the sensor
(converting it from analogue to digital via the ADC if needed)

It compares the sensor data to a stored threshold value
— is a person within 1 metre?

If the value confirms someone is nearby, the microprocessor
sends a signal to the output device (the screen) to display
the welcome message

Exam Tip: Always structure a microprocessor answer as:
receive data → compare to stored value → send signal to output.
That earns all 3 marks.

Q27 — Identify Three Other Characteristics of AI [3 marks]

(Feb/Mar 2025 — 0478/12/F/M/25 Q6a)

One characteristic of AI is the ability to learn.
Identify three other characteristics of AI.

Model Answer (any 3 from the remaining 4)

Simulates intelligent human behaviour

Collection of data and rules

Ability to reason

Analyses patterns

Exam Tip: This question gives you one for free and asks for three more.
Do not write "ability to learn" again — you will get zero for a repeated point.
Write three short, clean, distinct points. One line each.

Q28 — Machine Learning in a Search Engine [3 marks]

(Feb/Mar 2025 — 0478/12/F/M/25 Q6b)

A search engine uses machine learning.
Explain how machine learning is used by the search engine.

Model Answer

The search engine collects data from millions of searches —
what users searched for, which results they clicked on, and
how long they stayed on each page

It uses this data to identify patterns — for example,
which results are most useful for a given search term

The search engine automatically adapts its own ranking
process over time, improving the results it shows without
anyone reprogramming it

Exam Tip: The word "explain" means you need to say HOW it works
— not just define machine learning.
Hit three things: collects data, finds patterns, adapts automatically.

Q29 — Robotics: Characteristics of a Robot [3 marks]

(0478/11/M/J/23)

A robot is used in a factory to pack boxes.
State three characteristics of a robot.

Model Answer

A robot has a mechanical structure — a physical body that can
move and interact with the world

A robot has electrical components — sensors, a microprocessor,
and actuators working together

A robot is programmable — it follows a stored set of instructions
that can be changed

Q30 — Robotics: Independent vs Dependent [3 marks]

(0478/11/M/J/23)

The factory also uses a robot arm that is controlled remotely
by a human operator.
State the type of robot this is and explain the difference
between this type and a fully autonomous robot.

Model Answer

This is a dependent robot — it requires a human to control it
through an interface and cannot operate on its own

A fully autonomous (independent) robot operates without any
human intervention — it makes all its own decisions

The key difference is that a dependent robot supplements the human
whereas an independent robot completely replaces the human role

Q31 — Robotics: Hospital Robot [4 marks]

(0478/11/M/J/24)

A hospital uses a robot to deliver medicines to patients in different
wards. Describe two advantages and one disadvantage of using
this robot instead of a human worker.

Model Answer

Advantages (any 2)

The robot works 24/7 without rest — medicines can be delivered
at any time of day or night without delays

The robot is more consistent and precise — it delivers exactly
the right dose to the right patient without human error

Disadvantage (any 1)

The robot is expensive to install and maintain — the hospital
must invest heavily upfront and pay for technical specialists

The robot cannot handle unexpected situations — if a patient
is in a different location than expected, it may fail to adapt

Q32 — Automated Systems: Flood Warning [4 marks]

(0478/11/M/J/24)

A river uses an automated system to detect dangerous flood levels
and trigger a warning siren.
Describe how this automated system works.

Model Answer

A water-level sensor continuously measures the height of the
river water

The data is analogue — it passes through the ADC to be
converted into digital binary before reaching the microprocessor

The microprocessor compares the water level reading to a
stored safe maximum value

If the water level exceeds the threshold — the microprocessor
sends a signal to the actuator (siren) to sound the alarm

If the level is within the safe range — no action is taken

The process repeats continuously — this is the feedback loop

Q33 — Automated Systems: Two Benefits of Using Sensors [2 marks]

(0478/12/O/N/23)

Give two benefits of using sensors in an automated system
rather than relying on human observation.

Model Answer

Sensors can detect changes much faster than a human can
notice and react — improving response time and safety

Sensors work continuously without rest — they never get
distracted, tired, or need a break, unlike a human observer

Q34 — Automated Systems: Role of the ADC [2 marks]

(0478/12/O/N/24 — Q on sensor data)

A temperature sensor in an automated greenhouse produces
an analogue output.
Explain why an ADC is needed before the microprocessor can
use this data.

Model Answer

The temperature sensor produces an analogue signal —
a continuously changing value with infinite possible levels

The microprocessor can only process digital (binary) data —
0s and 1s. The ADC converts the analogue signal into binary
so the microprocessor can read and compare it

Q35 — Robotics: Why Robots Cannot Replace Doctors [3 marks]

(0478/13/O/N/24)

A hospital uses a robot to help diagnose patients.
Explain why the robot cannot fully replace a human doctor.

Model Answer

The robot can only follow its pre-programmed instructions —
it cannot handle new or unusual symptoms it was not programmed for

A robot has no empathy or human judgement — it cannot pick up
on how a patient is feeling emotionally or read non-verbal cues

Without AI added to it, the robot cannot learn or adapt from
new cases — a doctor improves with every patient they treat

Q36 — AI vs Expert System vs Machine Learning [4 marks]

(0478/13/M/J/23 — Stretch Question)

A hospital uses an expert system and a machine learning program
to assist doctors.
(a) Explain one way the expert system and machine learning
program are similar.
(b) Explain one way they are different.

Model Answer

(a) Similarity

Both are examples of artificial intelligence — they both
simulate intelligent human behaviour and make decisions
based on stored knowledge and data

(b) Difference

An expert system uses rules written by human experts
— it cannot update those rules on its own

A machine learning program automatically adapts its own rules
by finding patterns in new data — no human has to reprogram it

Quick Revision Flash Cards

Word	What It Means
Automated System	A machine that works on its own — no human has to press buttons or decide
Sensor	A device that detects something in the real world — temperature, movement, light
ADC	Translates the real-world signal into 0s and 1s the processor can read
Microprocessor	The brain — compares data to stored values and decides what to do
Actuator	The muscle — does something physical like turning on a motor or opening a valve
Feedback Loop	The system keeps repeating: sense, compare, act, sense, compare, act...
Robot	A machine with a physical body, electrical parts, and stored instructions
AI	Computer software that acts in an intelligent, human-like way
Expert System	AI that acts like a professional expert in one specific subject
Knowledge Base	Where all the facts are stored in an expert system
Rule Base	Where all the IF-THEN rules are stored in an expert system
Inference Engine	The part that reads the facts, applies the rules, and gives an answer
Interface	Where the user types questions and reads answers in an expert system
Machine Learning	A type of AI that teaches itself by looking at examples — no human writes the rules

Glossary

Word	Simple Meaning
Analogue	Something that changes smoothly — like temperature drifting gradually upward
Digital	Only 0s and 1s — the only language computers understand
Feedback Loop	The system keeps checking and acting, over and over, non-stop
Deskilling	When people forget how to do something because a machine does it for them
Hazardous	Dangerous — like radioactive areas, toxic fumes, or extreme heat
Cyber-attack	When a criminal tries to break into or damage a computer system remotely
Inference	Reaching a conclusion by thinking through the facts and rules logically
Subset	A smaller category inside a bigger one — ML is a subset of AI

Exam Command Words

Word	What the Examiner Wants
State / Give	Just write the fact — no explanation needed, keep it short
Describe	Walk through it step by step — what happens, then what, then what
Explain	Say WHY something happens, not just WHAT happens
Discuss	Give the good points AND the bad points
Identify	Point something out from a diagram or description given to you
Compare	Give both similarities AND differences
Outline	A short summary — just the most important points

The SCAR Formula

Every time an exam question says "describe how an automated system works",
use this four-letter formula. It covers everything the examiner wants.

Every automated system answer needs four things:
1.
The specific sensor — name it and say what it detects
2.
The ADC — if the data is analogue
3.
What happens when the value is safe (nothing — no action)
4.
That the process is continuous / it is a feedback loop

End of Chapter 6 — You have got this.

Table of Contents#

6.1 Automated Systems#

How It Works — The Control Loop#

What Is the ADC and Why Do We Need It?#

Common Sensors — What They Detect#

Good Things and Bad Things About Automated Systems#

Where Automated Systems Are Used#

6.2 Robotics#

Three Things Every Robot Must Have#

Two Types of Robot#

Where Robots Are Used#

Good Things and Bad Things About Robots#

6.3 Artificial Intelligence#

Five Characteristics of AI#

6.4 Expert Systems#

The Four Parts of an Expert System#

How the Inference Engine Works — Step by Step#

The Four Parts — Easy to Remember#

Where Expert Systems Are Used#

Good Things and Bad Things#

6.5 Machine Learning#

Normal Programming vs Machine Learning#

Where Machine Learning Is Used#

AI vs Machine Learning — What Is the Difference?#

Past Paper Questions and Model Answers#

Q1 — Greenhouse Temperature [8 marks]#

Q2 — Underground Mine [4 marks]#

Q3 — Garden Sprinkler [6 marks]#

Q4 — Characteristics of AI [3 marks]#

Q5 — AI in a Robot [4 marks]#

Q6 — Two Parts of an Automated System [2 marks]#

Q7 — Robots in a Car Factory [3 marks]#

Q8 — Expert System Fill in the Blanks [3 marks]#

Q9 — Machine Learning [2 marks]#

Q10 — AI vs Machine Learning [3 marks]#

Q11 — Expert System Components [4 marks]#

Q12 — Self-Parking Car [4 marks]#

Q13 — ATMs vs Human Bank Tellers [5 marks]#

Q14 — Specimen Paper 1B Q10(a) — Characteristics of AI [3 marks]#

Q15 — Specimen Paper 1B Q10(b) — Expert System Components [3 marks]#

Q16 — Specimen Paper 1B Q10(c) — Machine Learning [1 mark]#

Q17 — June 2024 Paper 12 Q10(a) — Garage Expert System [3 marks]#

Q18 — June 2024 Paper 12 Q10(b) — How the Garage Expert System Works [6 marks]#

Q19 — Oct/Nov 2023 Paper — Describe How an Expert System Operates [6 marks]#

Q20 — Machine Learning Improving a Robot [3 marks]#

Q21 — Why Does an Expert System Need a Knowledge Base? [2 marks]#

Q22 — State the Difference Between an Expert System and Machine Learning [2 marks]#

Q23 — Advantages and Disadvantages of Expert Systems [4 marks]#

Q24 — Supervised vs Unsupervised Machine Learning [2 marks]#

Q25 — ATM Proximity Sensor [1 mark]#

Q26 — Role of Microprocessor in ATM Detection [3 marks]#

Q27 — Identify Three Other Characteristics of AI [3 marks]#

Q28 — Machine Learning in a Search Engine [3 marks]#

Q29 — Robotics: Characteristics of a Robot [3 marks]#

Q30 — Robotics: Independent vs Dependent [3 marks]#

Q31 — Robotics: Hospital Robot [4 marks]#

Q32 — Automated Systems: Flood Warning [4 marks]#

Q33 — Automated Systems: Two Benefits of Using Sensors [2 marks]#

Q34 — Automated Systems: Role of the ADC [2 marks]#

Q35 — Robotics: Why Robots Cannot Replace Doctors [3 marks]#

Q36 — AI vs Expert System vs Machine Learning [4 marks]#

Quick Revision Flash Cards#

Glossary#

Exam Command Words#

The SCAR Formula#

Table of Contents

6.1 Automated Systems

How It Works — The Control Loop

What Is the ADC and Why Do We Need It?

Common Sensors — What They Detect

Good Things and Bad Things About Automated Systems

Where Automated Systems Are Used

6.2 Robotics

Three Things Every Robot Must Have

Two Types of Robot

Where Robots Are Used

Good Things and Bad Things About Robots

6.3 Artificial Intelligence

Five Characteristics of AI

6.4 Expert Systems

The Four Parts of an Expert System

How the Inference Engine Works — Step by Step

The Four Parts — Easy to Remember

Where Expert Systems Are Used

Good Things and Bad Things

6.5 Machine Learning

Normal Programming vs Machine Learning

Where Machine Learning Is Used

AI vs Machine Learning — What Is the Difference?

Past Paper Questions and Model Answers

Q1 — Greenhouse Temperature [8 marks]

Q2 — Underground Mine [4 marks]

Q3 — Garden Sprinkler [6 marks]

Q4 — Characteristics of AI [3 marks]

Q5 — AI in a Robot [4 marks]

Q6 — Two Parts of an Automated System [2 marks]

Q7 — Robots in a Car Factory [3 marks]

Q8 — Expert System Fill in the Blanks [3 marks]

Q9 — Machine Learning [2 marks]

Q10 — AI vs Machine Learning [3 marks]

Q11 — Expert System Components [4 marks]

Q12 — Self-Parking Car [4 marks]

Q13 — ATMs vs Human Bank Tellers [5 marks]

Q14 — Specimen Paper 1B Q10(a) — Characteristics of AI [3 marks]

Q15 — Specimen Paper 1B Q10(b) — Expert System Components [3 marks]

Q16 — Specimen Paper 1B Q10(c) — Machine Learning [1 mark]

Q17 — June 2024 Paper 12 Q10(a) — Garage Expert System [3 marks]

Q18 — June 2024 Paper 12 Q10(b) — How the Garage Expert System Works [6 marks]

Q19 — Oct/Nov 2023 Paper — Describe How an Expert System Operates [6 marks]

Q20 — Machine Learning Improving a Robot [3 marks]

Q21 — Why Does an Expert System Need a Knowledge Base? [2 marks]

Q22 — State the Difference Between an Expert System and Machine Learning [2 marks]

Q23 — Advantages and Disadvantages of Expert Systems [4 marks]

Q24 — Supervised vs Unsupervised Machine Learning [2 marks]

Q25 — ATM Proximity Sensor [1 mark]

Q26 — Role of Microprocessor in ATM Detection [3 marks]

Q27 — Identify Three Other Characteristics of AI [3 marks]

Q28 — Machine Learning in a Search Engine [3 marks]

Q29 — Robotics: Characteristics of a Robot [3 marks]

Q30 — Robotics: Independent vs Dependent [3 marks]

Q31 — Robotics: Hospital Robot [4 marks]

Q32 — Automated Systems: Flood Warning [4 marks]

Q33 — Automated Systems: Two Benefits of Using Sensors [2 marks]

Q34 — Automated Systems: Role of the ADC [2 marks]

Q35 — Robotics: Why Robots Cannot Replace Doctors [3 marks]

Q36 — AI vs Expert System vs Machine Learning [4 marks]

Quick Revision Flash Cards

Glossary

Exam Command Words

The SCAR Formula