Automated and Emerging Technologies

• Say it as one sentence and the whole of 6.1 follows: the sensor measures, the microprocessor compares with the preset value and decides, the actuator acts, and the sensor measures again.
• Before you write any evaluation point, ask one question: could I write this exact sentence about a completely different system? If yes, it is not yet an answer. Add the sensor, the number, the cost or the consequence that makes it belong to this scenario only.
• A robot's advantages and disadvantages both come from the same fact: it has a body that does exactly what the program says. That is why it is tireless, repeatable and safe to send into danger — and equally why it is costly, needs guarding, wears out, and cannot cope when the world is not where the program expects it to be.
• Two sentences hold all of 6.3 apart: an expert system reasons with rules a person wrote down; machine learning adapts a model it derived from data. Both are AI, because both simulate intelligent behaviour. Neither understands anything.
• If you remember nothing else: the sensor measures, the microprocessor compares with the preset value and decides, the actuator acts, and then the sensor measures again. Robotics adds a body you can point at. AI adds reasoning — and the word simulation is a limit on the claim, not decoration.

• Define an automated system and say what makes a system automated rather than merely electrical.
• State the role of a sensor, a microprocessor and an actuator, and say which are inputs and which are outputs.
• Describe, in order, how those three components collaborate: measurement, input data, comparison with a preset value, decision, output signal, physical action, repeat.
• Explain why a monitoring system may have no actuator at all, and identify when a scenario requires one.
• Apply the control cycle to an unfamiliar scenario and name the specific sensor, property, decision and actuator involved.
• Describe automated systems in industry, transport, agriculture, weather, gaming, lighting and science.
• Give advantages and disadvantages of an automated system that are justified by the given scenario, each with a reason and a consequence.
• Define robotics, and give the Cambridge examples of robots.
• Describe the three characteristics of a robot and explain what each one contributes.
• Describe roles robots perform in industry, transport, agriculture, medicine, domestic settings and entertainment, and evaluate their use in each.
• Define artificial intelligence and explain what “simulation of intelligent behaviour” does and does not claim.
• Describe the main characteristics of AI, using the syllabus wording that learning and adapting are things AI can include.
• Name and describe the four components of an expert system, and trace how a conclusion is reached.
• Define machine learning and describe, at concept level, how a program adapts its own processes or data.
• Compare an expert system with machine learning, and distinguish automated systems, robots and AI.

Why bias matters in a syllabus answer. Suppose a fruit-sorting model is trained only on apples grown in one region, photographed under one kind of light. Deployed at a packing house that receives fruit from elsewhere, it will misclassify more often — not because it broke, but because the new cases do not resemble what it learnt from. That single example covers three examinable points at once: output is not guaranteed, training data shapes the model, and more data would only help if it were representative data.

Microprocessor

The processing component of an automated system; it processes sensor data, compares it with a stored preset value or range and decides, according to its stored program, whether to send an output signal to an actuator.

Control Systems in Industry

Control systems in industry, transport and agriculture are automated systems in which sensors measure a physical property, a microprocessor compares each reading with a stored preset value, and actuators are switched to change the conditions being measured.

Machine Learning

An area of artificial intelligence in which a program has the ability to automatically adapt its own processes and/or data, building an internal model from example data and using it to classify, predict or decide.

Characteristics of a Robot

The characteristics of a robot are a mechanical structure that lets it move, electrical components such as sensors, motors and a power supply, and programmability, meaning it follows a set of stored instructions that can be changed.

Rule Base

The component of an expert system that stores the rules, commonly represented as IF-THEN rules, which state what conclusion follows from a given combination of facts.

Feedback Loop

The repeating cycle in which an actuator changes a physical condition, the sensor measures the changed condition, and the microprocessor uses that new reading to decide the next action.

Automated System

A system in which sensors, a microprocessor and — where the task requires them — actuators work in collaboration so that a process is monitored, and often controlled, according to programmed instructions with limited or no continuous human intervention.

Preset Value

A target value, limit or range stored in the microprocessor against which each sensor reading is compared, so that the system can decide whether an action is required.

Evaluating the Use of Robots

Evaluating the use of robots is the weighing of their benefits and drawbacks against a stated application area, where each point is argued from what that particular task demands rather than offered as a general statement about robots.

Robotics

A branch of computer science that incorporates the design, construction and operation of robots; examples of robots include factory equipment, domestic robots and drones.

Sensor

An input device that detects or measures a physical property such as temperature, light, moisture or pressure and sends the measurement to the microprocessor as input data; it takes no decisions.

Knowledge Base

The component of an expert system that stores the facts and domain-specific expert knowledge about one narrow subject area, gathered from human experts.

Actuator

An output component that receives a signal from the microprocessor and converts it into a physical action such as movement, heat, flow, light or sound; examples include motors, pumps, valves, heaters, buzzers and lamps.

Inference Engine

The reasoning component of an expert system; it applies the rules in the rule base to the facts supplied by the user, drawing on the knowledge base, and works towards a conclusion, generating further questions where information is missing.

Evaluating an Automated System

Evaluating an automated system is the weighing of its advantages against its disadvantages using the particular scenario given, where each point is stated with a reason and the consequence it produces in that setting rather than as a general claim.

Monitoring Systems

A monitoring system is an automated system in which sensor readings are recorded, displayed or used to raise an alarm, but no actuator changes the conditions being measured — which is what separates it from a control system.

Roles of Robots

The roles of robots are the specific tasks they carry out across the syllabus areas of industry, agriculture, medicine, domestic use, entertainment and transport, described by what the robot physically does rather than by naming the sector it works in.

Robot

A machine having a mechanical structure or framework, electrical components such as sensors, microprocessors and actuators, and the ability to be programmed to carry out a task.

Artificial Intelligence

A branch of computer science dealing with the simulation of intelligent behaviours by computers; its characteristics are the collection of data, the rules for using that data, the ability to reason, and it can include the ability to learn and adapt.

Expert System

An artificial intelligence system that reproduces the decision-making of a human expert within one narrow domain, built from a knowledge base, a rule base, an inference engine and an interface.

• M1. “Every automated system is artificially intelligent.” Why wrong Almost every automated system in this syllabus compares one measurement with one preset value and switches something. That is rule-based control, and no reasoning across a body of knowledge takes place. Say instead “The microprocessor compares the reading with the stored preset value and, following its stored program, sends an output signal to the actuator.”
• M2. “Every automated system contains a robot.” Why wrong Most have no mechanical body at all. A relay switching a lamp, a valve opening and a heating element are actuators, not robots. Say instead “The actuator is a switching relay, which turns the lamp on. There is no mechanical framework, so this is not a robot.”
• M3. “The sensor controls the actuator directly.” Why wrong This removes the comparison and the decision, which is where most of the marks in a process question sit. A sensor measures and reports; it has no route to the actuator. Say instead “The sensor sends the reading to the microprocessor, which compares it with the preset value and then sends the output signal to the actuator.”
• M4. “The microprocessor measures the temperature.” Why wrong A microprocessor never touches the environment. It receives a reading that a sensor produced. Say instead “The temperature sensor measures the temperature and sends the reading to the microprocessor as input data.”
• M5. “An actuator is an input device.” Why wrong It is an output component. Data flows one way: sensor to processor to actuator. Nothing is measured on the way back. Say instead “The actuator receives the output signal and produces the physical action. If the system needs to know the effect, a sensor measures the changed condition.”
• M6. “Automated systems are always correct.” Why wrong The system acts faithfully on whatever the sensor reports. A blocked, dirty, drifting or broken sensor still produces a number, and the system will act on that number as though it were true. Say instead “If the moisture probe corrodes and reads high, the microprocessor will conclude the soil is wet and will not irrigate, even though the crop is dry.”
• M7. “Automated systems never need human supervision.” Why wrong Someone sets the preset values, calibrates and maintains the sensors, and must be able to override the system when something happens that its rules do not cover. Say instead “Automation removes the need for continuous supervision, not the need for people.”
• M8. “Every automated system needs internet access.” Why wrong A sensor, a microprocessor and an actuator wired together form a complete automated system. A network is needed only if readings must travel somewhere. Say instead “The greenhouse controller works entirely locally. The weather station needs a transmitter because the readings must reach the forecasting centre, not because it is automated.”
• M9. “Every automated system must have an actuator.” Why wrong A monitoring system stores, displays or transmits data. A weather station is fully automated and changes nothing about the weather. Say instead “No actuator is required. The output of this system is the stored and transmitted record.”
• M10. “Automation always saves money and always causes unemployment.” Why wrong Both are overstatements. Automation trades a large initial cost and ongoing maintenance against lower running costs, and it changes which work is needed rather than simply removing work. Say instead “Running costs fall because water is applied only when needed, but the sensors and controller must be bought first, and it may take several seasons before the saving covers that.”
• M11. “Every robot is humanoid.” Why wrong The syllabus's own examples are factory equipment, domestic robots and drones — a bolted-down arm, a disc and an aircraft. Shape is not part of the definition. Say instead “It has a mechanical framework, electrical components and a stored program that can be changed, so it satisfies all three characteristics.”
• M12. “Every robot is autonomous.” Why wrong Autonomous means deciding its own actions without external control. Many robots follow a fixed stored sequence, and some are directed by an operator throughout. Say instead “The arm follows a stored sequence of coordinates. It is programmable, which is the characteristic required — it is not autonomous.”
• M13. “Every robot uses AI.” Why wrong A robot that repeats stored coordinates contains no reasoning and no learning at all. AI is a separate test. Say instead “The welding arm is a robot but contains no AI: it neither reasons across a body of knowledge nor adapts its own processes or data.”
• M14. “Any machine that uses electricity is a robot.” Why wrong A kettle, a lamp and a hairdryer are electrical. None has a programmable mechanical framework with sensors and actuators. Say instead “A washing machine is an automated system, but it is not a robot: its structure is fixed for one task and it cannot be reprogrammed to do another.”
• M15. “Robotics and AI mean the same thing.” Why wrong Robotics is a branch of computer science covering the design, construction and operation of physical machines. AI is the simulation of intelligent behaviour, and needs no body. Say instead “An expert system is AI but is not robotics; a fixed-sequence welding arm is robotics but is not AI.”
• M16. “A robot performs the surgery.” Why wrong In robot-assisted surgery the surgeon controls every movement from a console in the same room, and the team can convert to a conventional method if the equipment fails. Say instead “A robotic arm holds and moves the instruments under the direct control of a surgeon, whose hand movements can be scaled down and whose tremor is filtered out.”
• M17. “All AI systems learn and adapt.” Why wrong The syllabus says AI can include the ability to learn and adapt. An expert system reasons but does not normally learn, and it is still AI. Say instead “Its characteristics are the collection of data, the rules for using that data and the ability to reason; it can also include the ability to learn and adapt.”
• M18. “AI output is always correct, and always unbiased.” Why wrong An expert system is limited by its knowledge, its rules and what the user typed. A machine-learning model reproduces whatever is in its training data, including its errors and its bias. Say instead “The conclusion is a recommendation for a person to check, and its quality depends on the completeness of the knowledge base and the accuracy of the answers supplied.”
• M19. “An expert system and machine learning are basically the same thing.” Why wrong They differ in where the decision-making comes from: an expert system applies rules a person wrote down; machine learning applies a model the program derived from data. Say instead “In an expert system a specialist writes the rules into the rule base. In machine learning the program finds patterns in example data and builds the model itself.”
• M20. “The knowledge base and the rule base are the same component — and the inference engine stores the knowledge.” Why wrong Three separate components are being merged. The knowledge base holds facts; the rule base holds the rules that draw conclusions from facts; the inference engine holds neither — it is the reasoning process that consults both. Say instead “The inference engine searches the rule base for rules whose conditions the supplied facts satisfy, and applies them using knowledge from the knowledge base.”
• M21. “An expert system has common sense, so it will notice if a question makes no sense.” Why wrong It holds one narrow domain and nothing else. Outside that domain it has no knowledge and no basis for judging anything. Say instead “The system is limited to its domain. Given a symptom it has no rule for, it may produce a confident but wrong recommendation, which is why the output should be checked by a person.”
• M22. “Machine learning means the program rewrites its own code — and more training data always makes it better.” Why wrong What changes is the model — stored values, weightings and structures. The source code is unchanged. And more data helps only if it is accurate, relevant and representative; more of the wrong data makes a model more confidently wrong. Say instead “The program automatically adapts its own processes and/or data. Adding data helps only if the new data represents the cases the model will actually meet.”
• M23. “It recognised the pattern, so it understands — it might even be conscious.” Why wrong AI is defined as the simulation of intelligent behaviour. Output that resembles intelligence tells you about the output, not about anything happening inside the machine. Say instead “The system classified the image correctly. That is a computation over a learned model; it is not evidence of understanding.”

• The scope limit is a gift, not a restriction. The syllabus says AI here is limited to expert systems and machine learning. Writing three paragraphs about neural networks or chatbots in an answer about AI components does not gain credit and uses time you need elsewhere. Anything in this chapter marked Beyond the syllabus is context only.
• How many steps should you actually write? Match the marks. A 3-mark question wants roughly three linked steps — sensor measures, microprocessor compares with preset, actuator acts. A 5- or 6-mark question wants the fuller chain including the input data, the decision, the output signal and the repeat. Never write a bare list of three components for a 6-mark question: three nouns cannot earn six marks.
• “Discuss” and “evaluate” want a conclusion. Listing three advantages and three disadvantages is the body of the answer, not the end of it. A short final judgement — which side wins in this scenario, and on what condition — is where the AO3 marks live. Notice that the conclusion above names the condition that would reverse it.
• Read the bottom three rows again. The categories overlap, and a question that asks “is this a robot?” is not asking you to rule out the other two. The strongest answers say which tests are passed and which are not, and give the evidence for each.
• A quick self-test on all twenty-three. Cover the corrections. Read each myth aloud and, without looking, say the one word that makes it false: rule-based (M1), relay (M2), microprocessor (M3), sensor (M4), output (M5), calibration (M6), continuous (M7), local (M8), monitoring (M9), trade-off (M10), characteristics (M11), programmable (M12), separate (M13), framework (M14), body (M15), surgeon-controlled (M16), can (M17), training data (M18), derived (M19), consults (M20), domain (M21), model (M22), simulation (M23).
• Command words, decoded for Topic 6. These meanings are taken from the syllabus command-word table. What each command word wants when the subject is Topic 6 Command wordSyllabus meaningWhat that means hereWhich frame Stateexpress in clear termsOne short factual point per mark. Do not explain.none needed Identifyname / select / recogniseName the sensor, the actuator, the component. One or two words each.none needed Giveproduce an answer from a given source or from recallUsually an example. Make it a specific one.none needed Describestate the points of a topic / give characteristics and main featuresFor a process, the ordered chain. For a robot, the three characteristics with what each contributes.Frame 1 or 3 Explainset out purposes or reasons / make relationships clear / say why or howEvery point needs a “because” attached. Explaining that something happens is not explaining it.Frame 2 or 3 Compareidentify / comment on similarities and/or differencesMatched pairs in the same sentence, on the same dimension.Frame 4 Suggestapply knowledge to a situation with a range of valid responsesThe scenario's own details must appear in your answer. This is the command word that most punishes memorised lists.Frame 2 Evaluatejudge or calculate the quality, importance, amount or value of somethingPoints on both sides, then a stated judgement with the condition it rests on.Frame 2, then a conclusion Two words this chapter uses that are not in that table. The 2210 command-word table does not list Discuss or Justify. Where the practice questions in this chapter tag a question “discuss” or “justify”, treat them as this chapter’s own labels: discuss is being used in the sense of Evaluate — points on both sides and then a judgement — and justify in the sense of Explain — give the evidence for the decision you have just made. Answer them with the frame the table gives for Evaluate and Explain.