Lesson 6 · Differential Drive & Motor Control

Module 1 · Learning to Drive

Lesson 6 · Differential Drive & Motor Control

50–60 minPhase B · Driving ChallengesUnder the hood

👩‍🏫

Teacher mode is on. Toggle it off (bottom-right) to preview the student view.

Every Straight and Turn you've used is really controlling two motors behind the scenes. In this lesson you'll take the wheel yourself — driving the left and right motors directly to make curves, spins, and smooth steering that the high-level blocks can't do on their own. Along the way you'll find out what the word effort actually means, and why it isn't the same thing as speed.

Learning Objectives

By the end of this lesson you will be able to:

• Explain differential drive — steering by running the two wheels at different speeds
• Predict the robot's motion from what each side's motor is doing
• Use Set effort to control the motors directly (-1 to 1), and explain how effort differs from speed
• Decide when to use direct motor control vs. the high-level Straight/Turn

📄

Printable knowledge checks — the 8 questions from this lesson on one page, with room to write. Teachers: answer key.

How the robot really steers

The XRP is a differential drive robot: it has no steering wheel. The only thing it can control is how fast each side turns, and in which direction — and that turns out to be enough to produce every motion it can make.

Predict first: which way will the robot drive?

In each photo below, the robot is seen from above and the arrows show what each side's motor is doing: which way it turns, and how fast (longer arrow = faster). For each one, decide before you play the video: does the robot drive straight, curve, swing around, or spin? Then watch and check.

Case 1 — both sides forward, same speed. Which way will it drive?

Case 1 — press play once you've committed to a prediction.

Case 2 — both sides forward, but one side faster than the other.

Case 2 — press play once you've committed to a prediction.

Case 3 — one side driving, the other side stopped.

Case 3 — press play once you've committed to a prediction.

Case 4 — the two sides driving in opposite directions.

Case 4 — press play once you've committed to a prediction.

The whole rule, in three lines

Every one of those outcomes comes from the same idea — you operate the left and right sides of the driveline independently:

Same

Both sides, same speed

Drives straight. Match the two sides and there's nothing to turn the robot.

Different

One side faster

Turns. The bigger the difference, the tighter the curve — until one side stops and it swings about that wheel.

Opposite

Each side opposite

Turns in place. The robot rotates about its own center without going anywhere.

Knowledge Check

Your robot is supposed to drive straight but keeps drifting to the left. What does that tell you about the two sides?

Commanding the motors: the Set effort block

The Set effort block gives you that direct control. Effort runs from -1 (full reverse) through 0 (stopped) to 1 (full forward), and you set each wheel separately — so each of the four cases above is just a pair of effort numbers.

There's a catch, and it's the big idea of this lesson. Straight finishes: it drives its 20 cm, stops the motors, and hands the program to the next block. Set effort doesn't finish — it switches the motors on and moves straight to the next block, leaving them running. They keep running until something stops them.

So a Set effort program has to answer a question Straight never asks you: when should the motors stop? For a first look, we'll answer it with a Sleep — but keep an eye on how well that works.

← left: 0.5, right: 0.5 → straight

← 3 seconds — the motors are still running the whole time

Direct motor control: Set effort starts both wheels, and they stay on while the program sleeps.

Knowledge Check

You run Set effort with left 0.5 and right 0.5. What does the robot do?

Knowledge Check

What happens with left 0.5 and right -0.5 (one wheel forward, one backward)?

Effort is not the same as speed

The block is called Set effort, not "set speed," and the difference matters. Effort is how hard the motors try — electrically, it's how much voltage they get. Think of it the way you'd think about your own effort in a sport:

Effort is how hard you're working — not how fast you're actually going.

Those two things come apart as soon as the world pushes back. Pedal a bike uphill at a constant effort and you slow down; the same effort on the flat carries you much faster. Effort is your input. Speed is the result — and it depends on the hill, the carpet, the battery, and how much the robot is carrying.

Speed is the other way of asking: you name the speed you want, and the robot adjusts its own effort to hold it — pushing harder up the hill, easing off on the way down:

Ask for a speed and the effort changes to maintain it: big effort uphill, little effort coming down.

This is why the same Set effort numbers don't always produce the same motion. Effort 0.5 on a smooth floor is quicker than effort 0.5 on thick carpet, and quicker again with a fresh battery. It's also why Straight and Turn are more repeatable than raw effort — they watch the robot's sensors and adjust, instead of blindly holding one voltage.

Knowledge Check

You set both motors to effort 0.5 and drive across a smooth floor, then across a thick carpet. What happens?

Knowledge Check

A robot is asked to hold a steady speed as it drives up and over a hill. What does its effort do?

Effort vs. speed — worth naming explicitly

Students will otherwise read "effort 0.5" as "half speed," and then be puzzled when the same program behaves differently on carpet, on a ramp, or with a tired battery. The bike-uphill analogy lands fastest: same effort, different speed. If you have a ramp or a carpet square handy, run one Set effort program on both surfaces back to back — thirty seconds of demo and the distinction sticks.

This is also the honest answer to "why do we even have Straight and Turn?" Those close the loop with sensors to deliver a requested distance or angle; Set effort is open-loop and gives you whatever the world allows. Module 2's line-following builds directly on this idea of adjusting effort based on what the robot senses.

Blocks that finish vs. blocks that don't

Run the program above three times: once with a fresh battery, once on carpet, and once on a smooth floor. Measure how far the robot goes each time.

Three seconds at effort 0.5 is not a distance. It's three seconds — and the distance you get out of it changes with the battery, the surface, the weight on the robot, even whether the floor slopes. That's the effort-isn't-speed idea from the first section, showing up as a practical problem.

Now compare the two kinds of block:

Finishes

Straight / Turn

Runs until the job is done — the distance driven, the angle turned — then stops the motors by itself. The next block doesn't start until it's finished.

Doesn't finish

Set effort / Arcade

Switches the motors on and moves to the next block immediately. The motors stay on until something stops them.

Three things can stop them: a Stop motors block , another Set effort (including effort 0), or the program ending. You don't need a stop block at the end of a program — the robot stops the motors when the program is over — but any time the motors should quit while the program keeps going, that's what Stop motors is for.

Which leaves the real question: what should the robot wait for before it stops? A Sleep is the crudest possible answer, and you just measured why. The answer you'll use for the rest of this course is to stop on something the robot senses — drive until the rangefinder says the wall is 10 cm away, until the reflectance sensor sees the line, until the encoder has counted far enough. Same Set effort, but the robot decides when it's done instead of the clock. That's where Module 2 goes next.

Knowledge Check

What's the difference between what Straight does and what Set effort does?

Knowledge Check

Why is 'Set effort 0.5, Sleep 3 seconds' a poor way to travel a set distance?

Straight vs. Set effort — the trade-off

This is the lesson's big idea: abstraction vs. control. Straight/Turn use the robot's sensors (encoders/IMU) for precise, repeatable distances and angles — easy, but limited to straight lines and pivots. Set effort is raw and time-based: less precise, but it can make smooth curves and creative motion the high-level blocks can't. Neither is "better" — it's about the right tool for the task. The distinction to drill is finishing vs. not finishing: Straight and Turn stop the motors themselves when the job is done; Set effort and Arcade just start them. Two habits to head off while you're here. First, don't let timed driving take root — the Sleep in these examples is a placeholder because we have no sensors yet, and it's worth having students measure the same program on two surfaces so they see the distance change. Real programs stop on what the robot senses (Module 2), or use Straight/Turn when a measured distance is what they want. Second, nobody needs a stop block at the end of a program — the robot stops the motors when the program ends; Stop motors is for stopping mid-program.

Smooth steering with Arcade

Setting two efforts by hand is fiddly. The Arcade block is friendlier — you give it a forward speed and a turn amount (like a game controller), and it works out the two motor efforts for you.

← speed: 0.5, turn: 0.3 → forward, curving right

← 3 seconds — again, a stand-in until we have sensors

Arcade combines speed and steering into one intuitive block.

Knowledge Check

When is the Arcade block a better choice than Set effort?

Differentiation

More support: give a table of effort pairs and their outcomes, and start with symmetric efforts (0.3/0.3) before curves; let students use Arcade first — it's more intuitive. More challenge: build a spiral by gradually increasing one side's effort across several Set effort blocks.

Activity · Drive a figure-eight

A figure-eight is two curves that bend opposite ways, joined in the middle. It is the shape that proves you really understand what the two numbers in Arcade do — and Straight and Turn cannot draw it at all.

1. One curve first. Use Arcade with a speed of 0.5 and a turn of 0.3, held long enough to come most of the way around a circle. Adjust the turn number until the circle is about a metre across: bigger turn, tighter circle.
2. Write down what worked — your speed, your turn, and roughly how long the robot ran. You need those numbers twice.
3. Mirror it. Add a second Arcade with the same speed and the opposite turn (-0.3 if the first was 0.3), held for the same length of time.
4. Run the pair. You should get two loops joined in the middle. If the second loop is a different size from the first, the two halves aren't matched — check that the numbers really are mirror images.
5. Challenge — make the two loops different sizes on purpose, so the path looks like a lopsided eight. Which number do you change, and which stays?

You've succeeded when the robot traces two joined loops and finishes near where it began, pointing roughly the way it started.

Running the figure-eight

Floor space is the limiting factor: about 2 m × 1 m per robot. If that's tight, have students drop the speed to 0.3 — smaller loops, same lesson.

This is also where timed driving bites, hard and usefully. Two robots with the same numbers will trace visibly different eights, and the same robot will differ between a fresh battery and a tired one. Name it: the only reason we're timing this is that the robot can't yet sense where it is. That's the thread Module 2 picks up.

Step 2 matters more than it looks — students who don't record their first curve end up guessing twice and never get a symmetric eight.

Real-world connections

Direct motor control is how machines that don't drive in straight lines work:

Mobility

Wheelchairs & rovers

Powered wheelchairs and Mars rovers steer by differential drive — exactly this left/right speed trick.

Gaming

Twin-stick control

Tank-style games map two sticks to two "tracks," the same idea as setting two efforts.

Heavy equipment

Skid-steer loaders

Construction "skid steers" turn by driving their left and right wheels at different speeds.

Wrap-up

• How does a differential-drive robot turn? (By running its wheels at different speeds.)
• What happens to the motors when a Straight block finishes? What about a Set effort block? (Straight stops them; Set effort leaves them running.)
• Why is timed driving unreliable, and what should a robot stop on instead? (Effort isn't speed — battery and surface change the distance; stop on a sensor reading.)
• When would you pick Straight/Turn over Set effort? (When you need precise, repeatable distances and angles.)

Assessment & next lesson

Formative: can students predict motion from a pair of efforts and explain the precision/control trade-off? Exit ticket: "Fill in efforts to spin the robot in place" (equal and opposite, e.g. 0.5 / −0.5). Preview: Lesson 7 is a synthesis challenge that pulls together loops, functions, parameters, and motor control — the capstone of the Blockly phase.

Resources

• XRP Code IDE
• XRP Robot User Guide