Chapter 6: Controller for a Robotic System
It is the last period of a warm afternoon, and the ceiling fans in the classroom are running at full speed. By four o’clock the air has cooled a little, but the fans keep spinning exactly as fast as before. They go on that way until Aarav gets up and turns the regulator down two steps.
Down the corridor, the air conditioner in the school office is doing something quite different. Nobody has touched it since morning. It cools the room, stops on its own, waits, and starts again a few minutes later. In the staff room the refrigerator behaves the same way — its motor hums for a while, falls silent, then hums again.
Three machines, all dealing with temperature. One waited for a person to act. Two managed on their own, because each of them can measure what is happening and change what it does.
A robot faces the same problem, only much faster. A robotic arm told to stop at a certain angle has to keep checking where it actually is and correct the difference many times every second. The part of the robot that does this checking and correcting is its controller, and that is what this chapter is about. Fig. 6.0-1 puts the three everyday machines and the robot side by side.
Diagram
Fig. 6.0-1 – four small panels in a row: (a) a ceiling fan with a wall regulator and a hand turning the knob, (b) a wall-mounted air conditioner in a room with a thermometer beside it, (c) a refrigerator with its door closed and a thermometer inside, (d) a robotic arm holding a raised position. Below all four panels, one wide arrow loop with four labels around it: “set a target”, “measure what is happening”, “compare”, “correct”.
As you read this chapter, you will find the answer to each of these five questions.
6.1 What a Control System Is
Think about carrying a glass of water across a room without spilling it. You do not walk with your eyes shut. You keep looking at the water, and the moment it tilts towards one side you slow down or straighten your hand. You are doing three things over and over: deciding what you want, checking what is actually happening, and correcting the difference.
Machines have the same problem. A pump has to keep a tank full. A geyser has to hold the water at a comfortable temperature. A robotic arm has to stop at a certain angle and stay there. In every one of these cases, something has to keep watching the machine and keep adjusting it. That something is a control system — an arrangement that regulates how a device or a machine behaves so that it reaches and holds the result we want.
Setpoint, error, correction
The value we want is called the setpoint. It is simply the target — 25 °C for a room, 90° for a robotic joint, four litres for a tank. The value the machine has actually reached is the actual value. The gap between the two is the error.
\[\text{error} = \text{setpoint} - \text{actual value}\]
Here setpoint is the value asked for and actual value is the value measured on the machine. Both must be in the same unit, so an error in degrees is compared with a target in degrees, never with one in centimetres.
A control system works by keeping this error small. It measures the actual value, compares it with the setpoint, and changes what the machine is doing so that the error shrinks. Then it measures again. This repeating path — measure, compare, correct, measure again — is called a control loop. Fig. 6.1-1 shows the shape of one loop.
Diagram
Fig. 6.1-1 – a loop drawn left to right: a box labelled “setpoint (the value we want)” with an arrow into a small circle marked with a minus sign; from the circle an arrow labelled “error” goes into a box labelled “control system”; from there an arrow goes into a box labelled “machine”; an arrow leaves the machine to the right labelled “actual value”. A return arrow branches off the actual value line, runs back below the boxes, and enters the circle from underneath, labelled “measured value”.
Why robots depend on this
A robot is a machine that has to act in a world that keeps changing. The floor may be uneven. A gripper may pick up a heavy box one minute and an empty one the next. A motor that is warm turns a little differently from a motor that is cold. No fixed set of instructions can allow for all of this in advance. What works instead is a control system that keeps measuring and keeps correcting.
A control system has one honest limit, and it is worth remembering from the start: it can only correct what it can measure. If nothing measures the angle of a joint, no control system can hold that angle steady, however good the rest of the design is. This is why control systems and sensors always appear together in a robot.
6.2 Control Systems in Daily Life
You do not have to visit a factory to see a control system. There are several in an ordinary home, and looking at three of them closely — a fan, a refrigerator and an air conditioner — shows something useful. All three deal with comfort. All three have a target value. But they do not all keep that target in the same way, and the difference between them is exactly the difference we will need later when we look at robots.
6.2.1 Regulating a Fan
A ceiling fan has one job: to move air. On a mild morning you want it turning slowly; on a hot afternoon you want it at full speed. The fan itself has no way of knowing which kind of day it is, so a regulator is fitted on the wall — a switch or a knob that lets a person choose the speed.
Turning the regulator changes the electrical supply reaching the fan motor, and the motor turns faster or slower to match. Older regulators do this in fixed steps, usually five, while newer electronic ones can give a smoother range. Either way the chain is short: you choose a setting, and the fan holds that setting. See Fig. 6.2-1.
Diagram
Fig. 6.2-1 – a left-to-right chain of three boxes: “person chooses a setting” → “regulator” → “fan motor”, with an arrow leaving the last box labelled “air movement in the room”. A small figure of a person stands beside the first box, with a dashed arrow running back from the air movement arrow to the person and then to the regulator box, labelled “the person judges whether the room feels comfortable”.
Notice where the judging happens. Nothing inside the fan measures the temperature of the room. If the afternoon cools down, the fan carries on at exactly the speed it was left at. The comparing is done by a person, who feels the room and reaches for the regulator again. The fan is regulated, but it is the person who closes the loop.
6.2.2 The Refrigerator
A refrigerator has a harder job. Food stays fresh only inside a narrow range of temperature — typically about 3 °C to 5 °C in the main compartment. That target has to be held all day, whether the door has just been opened, whether the kitchen is hot, and whether the shelves are full or nearly empty. Asking a person to keep checking would be hopeless.
So the refrigerator carries its own temperature-measuring part, called a thermostat. The thermostat is set to a target temperature. When the air inside becomes warmer than that target, the thermostat switches on the compressor, which drives the cooling. When the inside has cooled to the target, the thermostat switches the compressor off again. This is why a fridge hums for a while, falls silent, and later starts humming again.
6.2.3 The Air Conditioner
An air conditioner works on the same idea as the refrigerator, but the space it cools is a room, and the target is chosen by the person using it. You press a number on the remote — 24 °C or 26 °C on a typical setting — and that number becomes the setpoint.
A temperature sensor inside the indoor unit keeps measuring the air of the room. While the room is warmer than the setpoint, the compressor runs and cool air is delivered. Once the room reaches the setpoint, the compressor stops, though the indoor fan usually keeps turning. As the room warms up again — from sunlight, from people, from a door left open — the sensor detects the rise and the compressor starts once more. Fig. 6.2-2 shows this rise and fall.
Diagram
Fig. 6.2-2 – a line graph with time on the x-axis and room temperature on the y-axis. A horizontal dashed line across the middle is labelled “setpoint”. A wavy solid line dips just below the dashed line, rises just above it, and repeats, staying close to it throughout. Short labels on the wavy line read “compressor on” along the falling parts and “compressor off” along the rising parts.
The important point is that nobody is watching the room. Once the setpoint is chosen, the air conditioner measures, compares and corrects on its own, again and again, for as long as it is running.
Three machines, one comparison
Table 6.1 sets the three side by side. Read across the row that asks what measures the result: it is the machine in two cases, and a person in the third.
This difference — whether the machine measures its own result or leaves that to a person — is the idea the next section is built on.
6.3 Open Loop and Closed Loop Control Systems
The fan, the refrigerator and the air conditioner all reached a target, but they did not all do it in the same way. The fan held whatever setting it was given. The other two kept checking the result and adjusting themselves. That single difference splits control systems into two families, and almost every machine you meet belongs to one of them.
The word for what the refrigerator and the air conditioner have, and the fan does not, is feedback — information about the actual output of a machine, sent back so that it can be compared with the target.
When the machine does not look back
In an open loop control system, the control action does not depend on the output at all. The system is given a setting, it acts on that setting, and it never checks what actually happened. There is no path carrying the result back for comparison. Fig. 6.3-1 shows the shape: a straight line from one end to the other, with nothing returning.
Diagram
Fig. 6.3-1 – a straight left-to-right chain of three boxes with arrows between them: “input (the setting chosen)” → “controller” → “process or machine” → an arrow leaving to the right labelled “output”. No return path anywhere in the diagram; add the caption text “nothing comes back”.
A ceiling fan on a regulator is a good example. So is a simple room light on a switch, and a mixer set to speed 2. Each of them does exactly what the setting says. If the room cools, if the bulb dims, if the mixture thickens, the machine does not notice, because nothing is measuring the result and nothing is sending it back.
When the machine checks its own result
In a closed loop control system, the control action depends on the output. The output is measured, sent back, and compared with the setpoint, and the difference decides what happens next. The path from output back to the comparison is what closes the loop, which is where the name comes from. See Fig. 6.3-2.
Diagram
Fig. 6.3-2 – the same left-to-right chain, “input (setpoint)” → a small circle marked with a minus sign → “controller” → “process or machine” → arrow leaving to the right labelled “output”. A return arrow branches from the output line, passes through a box labelled “sensor” placed below the chain, and feeds back into the circle, labelled “feedback”. Mark the arrow between the circle and the controller as “error”.
The refrigerator and the air conditioner are both closed loop systems, and so is an automatic electric iron, which cuts the heating element off when the sole plate reaches the temperature chosen on the dial. In each case the machine measures, compares, and corrects without anyone standing over it.
Which is better?
It is tempting to say closed loop is always better, but that is not true, and knowing why matters. A closed loop system needs a sensor, needs somewhere to do the comparing, and costs more to build. For a job where nothing much can go wrong — switching a light on, running a fan — that expense buys very little.
Where closed loop earns its cost is where the result cannot be predicted from the setting alone. A robot arm carrying a heavy box sags more than one carrying an empty box, even with the same command sent to the motor. Only a measurement of the actual position can catch that difference.
Each family also has an honest weakness. An open loop system has no way of noticing a disturbance — if the belt slips or the load changes, the error simply stays. A closed loop system can notice, but it depends completely on its sensor: if the sensor reads wrongly, the system will confidently correct towards the wrong value. Table 6.2 sets the two side by side.
From here on, whenever a robot is described as correcting itself, keeping a position, or holding a speed, you are looking at a closed loop control system. The next section takes that loop apart and names its four pieces.
6.4 The Parts of a Control System
A closed loop can be drawn as a ring, but it is easier to work with once you know which piece does what. Whatever the machine — an air conditioner, a lift or a robotic arm — the same four parts appear: an input, a controller, the machine being controlled, and feedback. Engineers draw them as a block diagram, with each part in a box and arrows showing what passes between them. Fig. 6.4-1 is that diagram for a robot.
Diagram
Fig. 6.4-1 – a closed loop block diagram, left to right: a box labelled “input” with an arrow into a small circle marked with a minus sign; from the circle an arrow labelled “error” into a box labelled “controller”; from the controller an arrow labelled “control signal” into a box labelled “robot”; from the robot an arrow to the right labelled “output”. A return arrow branches from the output line, runs back below the boxes through an arrow labelled “feedback”, and enters the circle from underneath.
Read the diagram as a journey and it makes sense: a target goes in, the gap between the target and the measured result is worked out, the controller decides what to do about that gap, the robot moves, and the result of the movement is measured and sent back. Each of the next four parts is one box in that ring.
6.4.1 Input
The input to a control system is the value we want it to achieve — the setpoint. It is not the electricity that powers the machine and it is not the sensor reading; it is the instruction that says what “correct” means for this task.
The input can come from a person, as when someone chooses 24 °C on an air-conditioner remote. It can come from a program, as when a line of code tells a robotic arm to hold a joint at 90°. It can also come from another part of a larger system, as when a driver-assistance system in a car is asked to hold a speed of 40 km/h. In each case the rest of the loop exists only to make the actual value match this number.
6.4.2 Controller
The controller is the part that receives the error and decides what the machine should do about it. It takes in the target and the measurement, works out the difference, and produces a control signal — an instruction telling the motor or heater how hard to work and in which direction.
In a small robot the controller is usually a microcontroller, the same kind of board used for the circuits in earlier chapters. In a larger machine it may be a microprocessor or a dedicated control board. What matters is not the hardware but the job: comparing, deciding, and issuing a signal, many times a second.
6.4.3 The Robot or the Process
The third box is the machine that actually does the physical work — the part that turns a control signal into movement, heat or flow. In a robot this box is the robot itself: its motors, joints and wheels. In an air conditioner it is the compressor and the cooling system. The general name engineers use for this box is the plant, whatever the machine happens to be.
This is where the actuators of the system sit. The controller cannot move anything by itself; it only produces a signal. It is a motor turning a joint, or a compressor driving the cooling, that changes the actual value in the real world.
6.4.4 Feedback
Feedback is the measured output travelling back to be compared with the input, and in almost every system it comes from sensors. A position sensor on a joint reports the angle the joint has actually reached. A wheel encoder reports how fast the wheel is really turning. A temperature sensor reports the temperature the cabinet has really reached.
The controller subtracts this measured value from the setpoint. If the difference is zero, the machine is where it should be and the control signal holds steady. If the difference is not zero, the sign of the error tells the controller which way to push and the size of the error tells it how much correction is needed.
Feedback carries one warning with it. Everything the loop does rests on the sensor being right. A loose encoder, a dusty optical sensor or a wire coming away will send back a wrong reading, and the controller will correct towards that wrong reading with complete confidence. This is why feedback sensors on a robot are checked as carefully as the motors are.
6.5 Control Systems in Robotics
In a house, a control system holds a temperature. In a robot it has to hold a position, a speed, a direction and a grip force, often all at the same time, and it has to do it while the ground, the load and the surroundings keep changing. The parts are the ones you met in the last section. What is new is how fast the loop runs and how many jobs depend on it.
6.5.1 The Control Loop Inside a Robot
Picture a small wheeled robot crossing a school corridor. It has been told to move at a steady speed. On smooth tiles this is easy, but at the doorway it meets a slight slope, and a motor pushing the same amount as before will now turn more slowly. Nothing has gone wrong with the robot; the world has simply changed under it.
An encoder on each wheel measures how fast that wheel is really turning and sends the reading back to the controller. The controller compares the reading with the commanded speed, finds the error, and increases the drive to the motor until the wheel is back at the commanded speed. When the robot leaves the slope, the same loop notices the wheel running fast and eases the drive off again. Fig. 6.5-1 shows the loop in place on the robot.
Diagram
Fig. 6.5-1 – a side view of a small four-wheeled robot climbing a shallow ramp, with a block diagram drawn above it. The block diagram runs: “commanded speed” → circle with a minus sign → “controller board” → “motor and wheel” → “actual speed”, with a return arrow through a box labelled “wheel encoder” back to the circle.
The loop is not run once. It repeats, and it repeats quickly — a small robot controller typically runs its loop tens or hundreds of times every second. Each pass makes only a small correction, but because the passes come so fast, the robot appears to move smoothly rather than in jerks.
The steps above can be tried out on a computer without any hardware at all, by pretending the wheel obeys each correction. Doing this shows the one thing worth seeing: the error shrinking pass after pass.
Look at the error column. It halves every pass and the speed climbs towards 60.0 rpm without ever jumping past it. This is a simulation running on a computer, not a robot — but the pattern is the real one, and it explains why a robot settles onto its target instead of arriving at it in one leap.
6.5.2 What Control Systems Do in a Robot
Speed is only one of the things a robot has to hold steady. The same loop, wrapped around different sensors and different actuators, does several different jobs in a working robot.
Motion control is the job in the example above: keeping joints, wheels and grippers moving to the positions and speeds that were commanded. Encoders and servomotors do most of the work here, and it is what makes a robotic arm arrive at the same point every time.
Path planning decides the route. Sensors such as ultrasonic sensors, cameras and lidar report what lies ahead, the controller works out a path to the target that avoids the obstacles, and the motion control loop then carries the robot along it. When something new appears in the way, fresh sensor readings feed back and the path is worked out again.
Environmental interaction covers picking things up and putting them down. Tactile and pressure sensors on a gripper report how firmly it is holding an object, and the controller adjusts the grip — enough to lift a glass, not so much that it cracks.
Communication and coordination matters when several robots share a task, as a group of drones does when surveying a large field. Each one reports its position and receives instructions, so the group covers the area without two machines going to the same place.
Energy management is the quiet one. Feedback about how long a robot has been idle lets the controller shut down motors or put the machine into a low-power state, which matters a great deal for a robot running on a battery.
Across all these jobs the pattern never changes: measure the actual result, compare it with what was asked, correct the difference, repeat. What changes is which sensor supplies the measurement and which actuator makes the correction.
6.6 Manual and Automatic Control Systems
Sections 6.3 to 6.5 asked how a control loop is built. This section asks a different question: who is in the loop? In some robots a person is inside it, watching and steering all the time. In others the person has stepped out, and the machine runs the loop by itself. Fig. 6.6-1 shows the same wheeled robot under both arrangements.
Diagram
Fig. 6.6-1 – two panels showing the same wheeled robot in a school corridor. Left panel labelled “a person in the loop”: a student stands holding a remote, a dashed arrow runs from the robot to the student’s eyes labelled “the student sees where it is”, and a solid arrow runs from the remote to the robot labelled “steering commands”. Right panel labelled “the machine runs the loop”: the robot alone, with a distance sensor at the front, a dashed arrow from the sensor curving back into a small controller box drawn inside the robot, and an arrow from that box to the wheels.
6.6.1 Manual Control Systems
In a manual control system, a human operator directly controls the robot’s actions and movements through an input device such as a joystick, a keyboard or a remote control. The machine carries out the commands it is given, and nothing more.
The important point is where the measuring happens. The operator watches the robot — directly, or through a camera — judges what needs to change, and moves the joystick. The comparison and the decision take place in the person’s head, exactly as they did with the ceiling fan regulator in Section 6.2.1. This is why a manual system stays fairly simple: the machine does not need to work out what is going wrong.
Manually controlled machines are common wherever judgement matters more than repetition. A remote-controlled vehicle is one. A drone flown by an operator on the ground is another. A surgical robot is a third — the instruments move only as the surgeon moves the controls, and the machine adds steadiness rather than decisions.
Manual control has real strengths. In an unexpected situation the operator can decide what to do, which no program has to have anticipated. Its weakness is the operator: attention drifts, reactions take time, and a long or unreliable link between the controls and the robot makes fine work difficult.
6.6.2 Automatic Control Systems
In an automatic control system, the robot operates on pre-programmed instructions and algorithms, with little or no human interaction while it is running. A person sets the task and the limits; after that the machine reads its own sensors, compares, and corrects.
Automated guided vehicles that carry goods across a warehouse floor work this way, and so do the arms on an assembly line, service robots that clean floors, and drones that fly a survey route on their own.
What automatic control buys you is consistency. The machine does not tire and does not lose concentration, so the hundredth movement is made as precisely as the first. Because it acts on sensor readings rather than on a fixed script alone, it can also respond to changes around it — slowing when an obstacle appears, gripping harder when a load is heavier.
What it cannot do is handle a situation nobody wrote into it. Faced with something outside its program — a spill on the floor, a trolley left across its route — an automatic system may stop, or may keep acting on a reading that no longer means what the designer assumed. This is why serious installations keep an emergency stop within reach.
6.6.3 Comparing Manual and Automatic Control
In practice the two are not rivals, and most serious machines use both. A surgical robot is manually controlled but has automatic limits it will not let the surgeon cross. An automated guided vehicle runs by itself, yet a person can take over the moment something is wrong. Keeping a manual path available alongside an automatic one is a standard piece of safe design, not a sign of a weak system.
6.7 Summary
- A control system regulates the behaviour of a device or a machine so that its output stays close to the value we want.
- The value asked for is the setpoint, and the gap between the setpoint and the value actually reached is the error: error = setpoint − actual value.
- A control system works by repeating one path — measure, compare, correct — and this repeating path is called a control loop.
- A ceiling fan with a regulator changes speed only when a person moves the knob; nothing inside it measures the room.
- A refrigerator uses a thermostat to switch its compressor on when the inside becomes too warm and off when the set temperature is reached.
- An air conditioner measures the room temperature with a sensor in the indoor unit and cools until the set temperature is reached, so the room drifts slightly above and below the target.
- In an open loop control system the control action depends only on the input setting; the output is not measured and disturbances go uncorrected.
- In a closed loop control system the output is measured and fed back, so the error is found and reduced.
- Closed loop systems are more accurate but need more parts; an open loop system is the sensible choice when accuracy does not matter.
- A control system has four blocks: input, controller, robot or process, and feedback, drawn as a block diagram with the feedback subtracted from the input.
- The input is the desired value, the controller turns the error into a control signal, the robot or process carries out the action, and the feedback path measures the result using sensors.
- A robot’s control loop runs many times a second, easing the drive as the error shrinks; if the drive stays too strong the joint overshoots the setpoint and swings past it.
- Control systems in robots handle motion control, path planning, environmental interaction, communication and coordination, energy management, and autonomous decision making.
- In a manual control system a human operator controls the robot through a joystick, keyboard or remote, and the loop is closed through the person.
- In an automatic control system the robot follows pre-programmed instructions and its own sensor feedback, working with minimal or no human interaction while it runs.
- Manual control brings human judgement to unexpected situations; automatic control brings consistency and precision. Many real machines use both, with a manual path kept available for safety.
6.8 Key Terms
Automatic control system: A control system in which the machine follows pre-programmed instructions and its own sensor feedback, working with minimal or no human interaction.
Autonomous decision making: The controller selecting between the actions written into its program on the basis of the sensor readings it has just taken.
Closed loop control system: A control system that measures its own output and feeds that measurement back, so the control action is adjusted to reduce the error.
Communication and coordination: Control work in which several robots exchange information about their positions and actions so that they can share a task without colliding.
Control loop: The repeating path of measure, compare and correct that a control system follows.
Control system: An arrangement that regulates the behaviour of a device or a machine so that its output stays close to the value we want.
Controller: The part of a control system that receives the error and generates a control signal to drive the system towards the desired output.
Energy management: Control work that uses feedback about battery level and idle time to shut down parts of a robot that are not needed, so the charge lasts longer.
Environmental interaction: Control work in which touch and pressure feedback lets a robot hold or handle an object with the right amount of force.
Error: The gap between the setpoint and the value the machine has actually reached, found by subtracting the actual value from the setpoint.
Feedback: The measurement of what a system has actually achieved, carried back to be compared with the input.
Manual control system: A control system in which a human operator directly controls the machine’s actions through an input device such as a joystick, keyboard or remote control.
Motion control: Control work that uses feedback from the motors and joints to keep a robot’s movement smooth and to stop it in the right place.
Open loop control system: A control system in which the control action is decided only by the input setting, with no measurement of the output and no correction based on it.
Overshoot: Going past the setpoint because the drive was still strong when the target was reached.
Path planning: Control work that uses distance readings to work out and revise a route that reaches the target while keeping clear of obstacles.
Plant: The older name for the block that is being controlled — in robotics, the robot itself.
Setpoint: The value a control system is asked to reach and hold.
Thermostat: A temperature-operated switch that turns a machine such as a refrigerator on and off to hold the temperature that was set.
6.9 Objective Questions
Which of the following is an example of an open loop control system?
- A refrigerator with a thermostat
- An air conditioner set to 24 °C
- A mixer running at the speed selected on its switch
- A lift that stops level with the floor
Fill in the blanks: The value a control system is asked to reach is called the ____________, and the gap between that value and the value actually reached is called the ____________.
State whether the following statement is true or false: In an open loop control system, the output is measured and compared with the input before the next control action is taken.
In the block diagram Input → Controller → Robot → Feedback, which block produces the control signal?
- Input
- Controller
- Robot
- Feedback
Match each item in Column A with its role in Column B.
| Column A | Column B |
|---|---|
| a. Thermostat in a refrigerator | i. Feedback in the loop that fills the tank |
| b. Joystick used to drive a remote-controlled vehicle | ii. Controller |
| c. Position sensor on a robotic arm joint | iii. Input device in a manual control system |
| d. Level sensor fitted inside a water tank | iv. Feedback in the loop that holds the joint angle |
A robot joint has a setpoint of 60.0°. Its position sensor reports 57.5°. Write down the error at this moment, with its unit.
A controller keeps driving a joint motor at full strength even as the joint reaches its setpoint. What is the most likely result?
- The joint stops short of the setpoint
- The joint goes past the setpoint and swings back and forth
- The error becomes zero more quickly with no other effect
- The feedback path stops working
6.10 Concept Questions
Define a control system, and name the four blocks of its block diagram in the order in which a signal passes through them.
A ceiling fan with a regulator and an air conditioner both make a room more comfortable. Explain why only one of them is a closed loop control system.
Differentiate between a manual control system and an automatic control system, giving three points of difference.
In the block diagram of a closed loop control system, the feedback is subtracted from the input rather than added to it. Give the reason.
A robot wheel is asked to run at 0.50 m/s. On one pass through the control loop its speed sensor reports 0.38 m/s, and on the next pass it reports 0.46 m/s. Work out the error on each pass, and state what the controller does differently on the second pass and why.
The position sensor on a robotic arm joint comes loose and keeps reporting the same angle no matter how the arm moves. Explain what would happen to the joint, and state which honest limitation of control systems this shows.
6.11 Case-Based Questions
- A school fills its overhead tank with an electric pump. At present the caretaker switches the pump on and lets it run for a fixed 4 minutes, because the pump normally delivers 15 litres of water per minute. On a day when the mains pressure is low, the same pump delivers only 11 litres per minute.
- Name the type of control system the school is using at present, and state why it is of that type.
- Work out how much water the tank was expected to receive, how much it actually received on the low-pressure day, and the shortfall.
- Name one part the school could add to make this a closed loop system, and describe how the pump’s behaviour would change once it is fitted.
- A shopkeeper uses a small wheeled trolley robot to carry sacks from the storeroom to the counter. At present an assistant drives it with a remote control, watching it all the way. The shopkeeper is thinking of replacing it with a trolley that follows a marked route on its own, using a distance sensor at the front.
- Name the type of control system in use at present, and state who closes the control loop in it.
- Give two advantages the shopkeeper would gain from the new trolley, and one situation the new trolley might not handle well.
- Suggest one arrangement the shopkeeper should keep in place for safety once the trolley runs on its own, and give a reason for it.