Learn
A1.3.7 | REAL-WORLD CONTROL SYSTEMS (HL ONLY)
01 | HL: THE SAME IDEAS IN DIFFERENT MACHINES
Higher level only. A lift stops at a requested floor, a washing machine fills to a suitable water level and irrigation responds to dry soil. These systems perform different jobs, but each connects information, a decision rule and a physical action.
This lesson applies control-system components to all eight examples named in the specification: autonomous vehicles, home thermostats, automatic elevator controllers, washing machines, traffic signals, irrigation, home security and automatic doors.
To explain, go beyond naming a sensor and motor. Show what is measured, how the controller uses it, what changes physically and how later information affects the next decision.
Keep the boundary clear. A machine may contain several control loops, timed stages and safety interlocks. It is often inaccurate to label the whole machine simply open loop or closed loop without saying which action or variable you mean.
02 | A REUSABLE WAY TO EXPLAIN AN APPLICATION
Begin with the objective: what should the system achieve? Identify the relevant input or target, then describe a suitable measurement. Explain the decision rule and the actuator command before stating the physical result.
If the result is measured and used to adjust action, trace that feedback. Also identify disturbances or faults that could make the result differ from the command.
A safety interlock is a condition that prevents an action unless required conditions are met, such as not allowing lift travel without door-lock confirmation. An interlock input is important, but it is not automatically feedback about the main controlled variable.
A state describes the current stage of a sequence, such as filling, washing or draining. The same sensor reading can lead to different commands in different states because the controller is pursuing a different immediate objective.
| Question | Example detail to include |
|---|---|
| What is the objective? | Maintain a water level or reach a requested position |
| What information is supplied? | Target, sensor reading, user request or safety condition |
| What rule is applied? | Compare, select a state, enforce an interlock or calculate a command |
| What acts? | Motor, valve, heater, brake, lamp or alarm |
| What result returns? | Position, speed, temperature, level or traffic conditions |
| What can go wrong? | Faulty reading, actuator limit, delay or external disturbance |
03 | AUTONOMOUS VEHICLES: MANY CONNECTED CONTROL TASKS
An automated driving system must interpret its surroundings and control motion within its intended operating conditions. Cameras, radar, lidar where fitted, wheel-speed sensors and inertial sensors can supply different kinds of information. No single list applies to every vehicle design.
Perception and planning software estimate the situation and select a path or speed target. Lower-level controllers use measurements to command steering, propulsion and braking actuators. For example, measured speed can be compared with target speed, and wheel or inertial measurements help determine the response.
Feedback matters because a steering or braking command does not guarantee the intended trajectory. Changes in grip, slope, loading or obstacles can change the outcome, requiring updated estimates and commands.
Errors in sensing, delayed processing and limited braking or steering capability constrain performance. Driver-assistance features are not automatically full autonomy; the required human role and allowed conditions depend on the system. Explain a particular function, such as maintaining speed, rather than claiming a computer can safely handle every road situation.
04 | HOME THERMOSTATS: REGULATE A MEASURED QUANTITY
A user selects a target temperature. A temperature sensor reports a measurement to the thermostat controller, which applies a heating rule and commands the heating equipment through a suitable interface.
An on/off algorithm can use two thresholds: turn heating on below the lower threshold and off above the upper threshold. Between them it retains its previous state, reducing rapid switching.
The heater adds thermal energy, and later temperature readings report the result. This closes the temperature loop. An open window is a disturbance: increased heat loss can cause the controller to request more heating.
Feedback does not guarantee the target is reached. A biased sensor can cause the wrong actual temperature, while an undersized heater may be unable to offset heat loss. A timer that turns heating on for a fixed period without using temperature to adjust it is open loop with respect to room temperature.
05 | ELEVATORS: POSITION CONTROL PLUS A SEQUENCE
A passenger’s floor request is an input, while position sensors or an encoder provide information about the car’s location. The controller decides which request to serve and manages states such as closing doors, travelling, stopping and opening doors.
A drive motor moves the car, and braking equipment helps stop and hold it. Position and speed feedback support the required movement; simply powering the motor for a fixed time would not reliably reach the same floor under changing conditions.
Door sensors and lock confirmation provide interlock inputs. A controller must not start travel merely because it issued a close command: it needs the required confirmation that conditions permit motion. Doorway detection can prevent or reverse closure when an obstruction is present.
The floor-control loop and the door sequence interact, but are not identical. Position feedback answers “where is the car?” while the lock input answers “is the required door condition confirmed?” Real lift safety also involves additional engineered protection beyond this teaching model.
06 | EXPLORE A SIMPLIFIED ELEVATOR CONTROLLER
Choose a floor and step through the sequence. The motor may run only after closed doors and lock confirmation are established. A doorway obstruction prevents completion of closure. The model moves one floor per movement step and uses a position reading to stop at the target.
Use the buttons to simulate an obstruction or missing lock confirmation before travel. They are teaching inputs, not controls for a real lift. Inputs cannot be changed during travel in this model; real systems need additional responses to faults occurring while moving.
A step represents a state transition or observation, not a fixed duration. This is a conceptual control demonstration, not a complete lift controller.
Try removing lock confirmation before closing. The model waits for confirmation; it does not treat a command as proof that the doors are locked.
07 | WASHING MACHINES: TIMED STAGES AND FEEDBACK LOOPS
A selected programme defines a sequence such as filling, heating, washing, draining and spinning. The controller’s current state determines which sensor readings and rules matter at each stage.
During filling, a water-level or pressure sensor reports the level and an inlet valve controls flow. When the required level is detected, the controller closes the valve. Feedback helps compensate for differing inlet flow rates, unlike relying solely on a fixed filling duration.
A temperature sensor can control heating, and speed feedback can help regulate drum rotation. A door-lock input is an interlock; it does not itself measure water temperature or drum speed.
Some stages may use elapsed time, while others depend on measured conditions. Therefore a washing machine can combine open-loop timed actions, closed-loop regulation and state-based sequencing. Sensor failure, blocked drainage or unbalanced loads require appropriate handling; one fixed timer cannot describe the entire machine.
08 | TRAFFIC SIGNALS: FIXED PLANS AND RESPONSIVE CONTROL
A fixed-time signal follows a predetermined phase schedule. It may be open loop with respect to current traffic demand because it does not adjust that schedule using measured traffic conditions. A clock input alone is not evidence of traffic feedback.
A responsive controller can receive vehicle-detector information, pedestrian requests or measured traffic conditions. Within its rules, it selects or extends phases and commands signal lamps. Minimum times, clearance intervals and conflict rules constrain those decisions.
If measurements of traffic affected by the signals are used to adjust later operation, there is a traffic-responsive feedback relationship. A pedestrian button is a request; it does not by itself confirm that a lamp changed or that a crossing is clear.
The objectives include allocating movement opportunities and limiting unnecessary delay while maintaining required separation. Giving every road green immediately is not a valid response to high demand. Faulty detection and competing flows can limit the benefits of adaptive control.
09 | IRRIGATION: RESPOND TO SOIL CONDITIONS
A timer can open a watering valve on a schedule, but cannot alone tell whether the soil is already wet. A soil-moisture measurement gives the controller information directly relevant to the intended outcome.
A moisture-based rule can allow watering when a suitable reading indicates dry conditions and stop or inhibit it when sufficient moisture is present. A valve or pump changes water delivery, and later measurements reveal changes in the soil.
Rainfall, evaporation, soil type and plant uptake affect the result. Feedback can help avoid unnecessary watering compared with an unchanged time-only schedule, but sensor placement and measurement quality matter.
Water takes time to move through soil, so a controller may water in stages and allow time for measurements to respond. A weather forecast used to plan watering is predictive input, not direct measurement of the resulting soil moisture. Systems can combine both kinds of information.
10 | HOME SECURITY: DETECT EVENTS AND APPLY A RULE
A security controller can receive door-contact, motion or other detection signals alongside inputs such as armed mode and an authorised disarm command. Its rule determines whether an event should produce an alert.
For example, opening a protected door while armed may start an entry delay; if the system is not disarmed correctly, the controller can activate a siren and send a notification. The state and timing matter: the same door contact may not trigger an intrusion alarm when the system is disarmed.
A siren is an output transducer producing sound. A notification is a communication output, not proof that someone has acted on it. A door contact detects a condition; it is not a motor or the alarm algorithm.
Do not automatically call the intrusion rule a closed-loop siren controller just because it uses sensors. The door contact does not measure whether the siren actually sounded. A separate supervision signal could provide feedback about equipment status. False detections, tampering and power or communication faults must also be considered.
11 | AUTOMATIC DOORS: TRIGGER, MOVE AND VERIFY
A presence or approach detector can request opening, while a controller manages states such as closed, opening, open and closing. A motor actuator moves the door through a suitable mechanism.
A limit switch or position sensor can report that the door reached its open or closed position. That is outcome information used to stop or regulate motion, forming feedback about position. The approach signal has a different role: it requests or permits an action.
A doorway safety detector can inhibit or reverse closing while a person is present. When the doorway is clear, the controller may wait a specified time before closing. A timer does not replace the safety condition.
A fixed-duration motor command alone would be open loop with respect to actual position. Position feedback improves knowledge of the result, but jam detection, force limits and suitable fault handling still matter. A request to close is never the same as confirmation of safe completed closure.
12 | COMPARE THE ROLE OF THE INFORMATION
| Application | Request or target | Measurement / condition | Action |
|---|---|---|---|
| Vehicle speed control | Desired speed | Measured speed; relevant environment estimates | Adjust propulsion or braking |
| Thermostat | Desired temperature | Measured temperature | Command heating |
| Elevator | Requested floor | Position; door and lock conditions | Move, stop and operate doors |
| Washing machine | Programme and stage requirements | Water level, temperature, speed; interlocks | Operate valves, heater, drum and pump |
| Traffic signals | Policy, phase rules and requests | Traffic detection and relevant conditions | Change permitted signal phases |
| Irrigation | Watering objective and constraints | Soil moisture and other conditions | Open or close a valve; run a pump |
| Security | Armed/disarmed mode and authorised commands | Protected-zone events and equipment status | Alert, notify or apply configured actions |
| Automatic door | Opening request | Position and doorway safety conditions | Open, stop or close the door |
The sensor’s presence alone does not identify a feedback loop. State what outcome it measures and how that information affects subsequent action.
13 | OPEN LOOP, CLOSED LOOP OR MONITORING?
Classify each stated function rather than guessing from the whole device name. Monitoring means measuring or reporting without automatically adjusting the physical process in the example. A larger product may contain additional functions not described.
14 | EXPLAIN WHY FEEDBACK HELPS, AND WHERE IT DOES NOT
Feedback is useful when the same command can produce different outcomes. Increased lift load, reduced water pressure, heat loss or wheel slip can change how a physical system responds. Measuring the relevant result lets the algorithm respond to that difference.
However, the measurement may be delayed or inaccurate, and the actuator may be unable to supply the required action. A pump cannot deliver water from an empty supply simply because the algorithm requests it.
Several measurements can have different purposes. A washing-machine temperature sensor supports regulation, while a door-lock contact supports an interlock. Both are inputs, but their relationship to the objective differs.
A complete explanation identifies the purpose of each input and does not invent a feedback path. For an alarm, it is reasonable to describe event detection and conditional output without pretending the alarm restores the door to its previous state.
15 | WORKED EXPLANATIONS IN CONTEXT
Select a situation, form your explanation and then reveal a model answer. Each answer links the measurement to the rule, action and consequence rather than only listing hardware.
16 | EXPLAIN THE CAUSAL CHAIN
For explain, connect the components with reasons and consequences. A useful structure is: measurement or event → decision rule → actuator command → physical result → further information where applicable.
For irrigation: “The soil sensor reports moisture, so the controller can inhibit watering when the soil is already sufficiently wet. The valve therefore remains closed rather than following an unchanged time schedule, reducing unnecessary water use under those conditions.”
For an elevator: “The controller checks door-lock confirmation before permitting travel because a close command alone does not establish the required condition. Position measurements then allow it to determine progress towards the requested floor.”
Use because, therefore, resulting in and leading to when they clarify a real relationship. Do not force every application into a numeric setpoint loop: sequencing, event logic and interlocks are also important control behaviours.
17 | CHECK THE BOUNDARY AND THE LIMITATIONS
Before finishing an answer, check that your sensor measures what you claim, your actuator can cause the stated action, and the feedback path reports a relevant outcome. Identify which parts use time, which use requests and which use measured results.
Avoid claiming that all traffic lights use one fixed design, all sensors provide feedback about the final output, or every automated vehicle is fully autonomous. Real products vary; describe a plausible arrangement and state the assumptions.
This page gives conceptual examples, not complete safety designs. Real control systems require appropriate engineering, testing, fault handling and protection beyond the simplified rules shown here.
This completes the operating systems and control systems subsection. The next HL lesson moves to translation processes, including interpreters and compilers.