EV3 Robotics›Level 4 · WRO Prep›Lesson 31
Level 4 · Lesson 31 · EV3-L04-3160 minutes · Block 4 of 6 · Mat: WRO 2024 — Earth Allies, Sustainable Farming
The fastest settings that still finish ten runs out of ten.
Not the fastest lap you have ever driven. Not the fastest lap you drove today. The fastest settings you can prove complete every single time, ten times running, with nobody touching the robot.
This changes what tuning means. Every milestone before this asked whether the robot could do something. This one asks how often — and a setting that wins once and fails twice is worth less than a slower one that never fails.
The lap is your route from lesson 25 with the follower assignment from lesson 30: HOME at the loading yard, out to the stations you costed, and back to HOME.
Fix the lap now and do not change it today. The route, the missions, the order — all frozen. Today changes speeds and gains only. If the route moves halfway through, the ten runs are not ten runs of anything.
Decide before you start what counts as completed. Write it down. Something like: reached every station in order, the attachment did its job at each, returned to HOME, and nobody touched the robot. Agree it as a team beforehand, because agreeing it after a marginal run is not possible.
Every team, at some point today, will want to keep a fast setting that went nine out of ten. Here is the arithmetic that settles it.
A lap is made of legs, and the lap only works if every leg works. So the chances multiply.
A route of six legs, each of which “nearly always” works, fails about half the time. That is the difference between nine out of ten and ten out of ten, and it is much bigger than it feels.
Now run it the other way. Six legs at 99% each is about 94% — a clean lap almost every time. The gain from pushing a leg from 90% to 99% is far larger than the second you saved by driving it faster. That is the whole argument for reliability, in numbers.
And you get two rounds in a competition, not ten. A setting that works half the time gives you a fair chance of scoring nothing at all, on the day, in front of the judges.
Agree the procedure before any running, and follow it exactly. Its whole purpose is to be decided in advance, because after a failure every team can produce a reason the failure does not count.
One rule, and it is not negotiable: a failed run counts as failed. Not “that was the battery”, not “somebody bumped the table”, not “that one does not count”. On the day, none of those will count either.
Ten laps takes time. If your lap is long, run the marked fast leg ten times instead of the whole lap, and be clear in your journal that that is what the number covers.
When a set of ten fails, you get one change before the next set — same discipline as a mock round. Choose it from what you saw, not from what you wish were true.
Speeds are per leg, not global. After lesson 30 each leg has its own follower and its own speed, so the leg that fails is the leg that gets slowed. Slowing everything because one leg failed throws away time you had already earned.
Keep a running sheet as you go. Per settings tried:
That last column is the most useful thing on the page. Failures cluster. If seven of your last ten crosses are the same leg, you know exactly which leg is holding the whole lap back, and it is almost never the one that feels most dramatic.
When you have settings that went ten out of ten, run one final demonstration lap under the settings you are declaring, timed and watched.
That time is your milestone. Write it on the board next to the team name, with the note ten out of ten beside it.
It will probably be slower than the best single lap somebody drove earlier in the block, and that is worth saying out loud. The fast lap was luck plus a good run; this one is a claim you have evidence for. Only one of them survives a competition.
The milestone entry:
These settings are frozen for lesson 32. The mock round runs on what you certified today, and nothing gets retuned beforehand. That is the point — you are finding out whether a number proved in practice survives a round under pressure.
This model drives, so its challenges are run on a mat. Mats differ between branches — check you are looking at the one in your room.
WRO 2024 RoboMission Elementary — Earth Allies — Sustainable Farming · official WRO game mat, 2362 × 1143 mm
The challenges name these places rather than distances, so the same challenge works on any mat:
Switch mats above and every route below is redrawn on the mat you chose.
Work through the challenges in order — each is harder than the last. The mission comes after all three, and it is meant to make you plan before you build.
Freeze the lap — route, missions, order — and write down what counts as COMPLETED before any running: every station in order, the attachment doing its job at each, back to HOME, nobody touching the robot. Agreeing it after a marginal run is not possible.
Run the ratchet. Ten runs at a speed you are confident about; ten out of ten records the settings and raises the speed; the first failure stops you and you step back to the last settings that went ten. A failed run counts as failed — not the battery, not the table, not 'that one does not count'.
Keep a reliability sheet with a column for WHERE each failure happened. Failures cluster: if most of your crosses are the same leg, that leg is holding the whole lap back, and it is the only one that gets slowed.
Declare the fastest settings you can prove finish ten runs out of ten, run one final timed lap under them, and write the time on the board with 'ten out of ten' beside it. Put your best single lap next to it and note the gap — one of those is luck and a good run, the other is a claim you have evidence for, and only one survives a competition. Six legs at nine out of ten each is a 53% lap.