Automated and Emerging Technologies
How sensors, microprocessors and actuators work together to automate the world, what makes a robot, and the two faces of artificial intelligence: expert systems and machine learning. Topic 6 of the 0478 / 2210 syllabus in full.
The final Paper 1 topic looks at machines that run with little or no human help, and at the technologies that let computers appear to think. From an automatic greenhouse to a self-driving car to a medical diagnosis system, the same building blocks keep appearing. This chapter ties together the sensors and actuators you met in Chapter 3 and shows how they create automated and intelligent systems.
Automated Systems
An automated system uses sensors, microprocessors and actuators working together to monitor and control a process without continuous human intervention. The three components form a loop, and you must know the job of each.
Sensor
Measures a physical property such as temperature, light or pressure, and converts it into an electrical signal.
Microprocessor
Receives the sensor data, compares it to stored preset values, and decides what action to take.
Actuator
Receives the signal from the microprocessor and carries out a physical action, such as opening a valve or turning a motor.
Press play to watch the loop run. The temperature sensor reads a value, the microprocessor compares it to the target, and if it is too hot the actuator opens a window. The loop then repeats.
Automation Scenarios
The same sensor-microprocessor-actuator pattern appears across many fields. Being able to apply it to an unfamiliar scenario is exactly what the exam tests.
| Field | Example | How the loop applies |
|---|---|---|
| Agriculture | Automated irrigation | A moisture sensor reads the soil, and if it is too dry the microprocessor signals a valve actuator to water the crop |
| Transport | Automated traffic management | Sensors count vehicles, and the microprocessor adjusts traffic-light timing to ease congestion |
| Industry | Automated manufacturing | Sensors monitor a production line, and actuators move and assemble parts with consistent precision |
| Science | Climate monitoring | Sensors record conditions continuously, feeding data to a microprocessor for logging and alerts |
Advantages and disadvantages of automation
| Advantages | Disadvantages |
|---|---|
| Works continuously, 24 hours a day, without tiring | High initial setup and equipment cost |
| More precise and consistent than a human | Can replace human jobs, causing unemployment |
| Can operate safely in dangerous environments | Relies on maintenance and can fail or malfunction |
| Faster response and lower long-term running costs | Lacks human judgement in unexpected situations |
Robotics
Robotics is a branch of computer science involving the design, construction and operation of robots. A robot is a programmable machine with a physical structure that can carry out tasks automatically.
Characteristics of a robot
- A mechanical structure or framework: a physical body to perform tasks
- Electrical components: sensors (inputs), a microprocessor (processing) and actuators (outputs)
- Programmable: its behaviour is defined by software that can be changed
Where robots are used
| Area | Example | Benefit |
|---|---|---|
| Industry | Robotic welding arms on a car production line | Precision, speed and consistent quality |
| Medicine | Surgical robots | More precise than human hands, minimally invasive |
| Agriculture | Harvesting robots and crop-spraying drones | Can work continuously and reduce chemical use |
| Domestic | Robotic vacuum cleaners | Convenient and time-saving |
| Transport | Autonomous vehicles and warehouse robots | Safer and more efficient logistics |
| Entertainment | Animatronics and theme-park robots | Repeatable, lifelike performances |
Artificial Intelligence
Artificial Intelligence (AI) is a branch of computer science dealing with the simulation of intelligent behaviour by computers, making machines perform tasks that would normally require human intelligence.
Characteristics of AI
- Data collection: AI systems gather large amounts of relevant data
- Rules for using the data: defined rules or learned patterns decide how the data is processed
- The ability to reason: it can draw conclusions and make decisions from the information available
- The ability to learn and adapt: it can improve its performance over time as it receives new data
Expert Systems
An expert system is an AI program that mimics the decision-making ability of a human expert in a specific field, such as diagnosing an illness, using a store of knowledge and a set of rules.
The four components
Knowledge base
A large store of facts and information about the subject, for example medical symptoms, diseases and treatments.
Rule base
A set of IF-THEN rules that say how to use the knowledge, for example "IF fever AND rash THEN consider measles".
Inference engine
Applies the rules to the knowledge base to reason towards a conclusion from the inputs given.
User interface
The layer the user interacts with: it collects inputs such as symptoms and displays the conclusions.
How an expert system works
- The user enters information, such as symptoms, through the user interface
- The inference engine takes that input and applies the IF-THEN rules from the rule base to the facts in the knowledge base
- It reasons towards one or more conclusions, sometimes asking follow-up questions
- The conclusion, such as a likely diagnosis, is shown to the user through the interface
Machine Learning
Machine learning is a type of AI where a program automatically adapts its own processes or data based on experience. It learns from data rather than being explicitly programmed for every possible situation.
| Traditional programming | Machine learning |
|---|---|
| The programmer writes explicit rules | The system is trained on large datasets |
| The computer follows those rules exactly | It finds patterns in the data without explicit rules |
| Cannot handle situations the rules do not cover | Improves its accuracy as it receives more data |
| An expert system is an example | Examples: spam filters, image recognition, recommendation engines |
Exam Practice
A temperature sensor measures the temperature and sends the reading to the microprocessor. The microprocessor compares the reading with a stored preset value. If the temperature is above the preset value, the microprocessor sends a signal to an actuator, which opens a window or vent to cool the greenhouse. This process repeats continuously.
Advantages: it can work continuously without tiring, and it is more precise and consistent than a human. Disadvantages: it has a high initial setup cost, and it can replace human workers, leading to unemployment. Other valid points include working safely in dangerous conditions, and the system relying on maintenance and being able to malfunction.
A robot has a mechanical structure or framework, it has electrical components such as sensors, a microprocessor and actuators, and it is programmable, meaning its behaviour is controlled by software that can be changed.
The knowledge base stores facts and information about the subject. The rule base holds the IF-THEN rules for using that knowledge. The inference engine applies the rules to the knowledge base to reach conclusions. The user interface collects inputs from the user and displays the conclusions.
An expert system uses a fixed set of human-written IF-THEN rules applied to a knowledge base, and it does not improve by itself. Machine learning has no fixed rules: it is trained on large datasets, finds patterns in the data on its own, and improves its accuracy as it receives more data.
Any two of: it collects relevant data, it has rules or learned patterns for using that data, it has the ability to reason and draw conclusions, and it has the ability to learn and adapt over time.
