banner I have been a Lego League coach since 2007. This year, I wanted to document the season to give rookie coaches a resource to help them through to competition. The process can be intense, but it can also be a lot of fun for you and your team.

I hope to cover enough through my posts, but if I leave anything out, please feel free to leave a comment, or contact me.
Oct
27th

Wider is Better

Author: fllCoach | Files under Programming

When I heard this the other night, I was reminded of the car commercials that used to say “Wider is Better.” In their case, they were talking about car stability. In the case of robots, we’re talking accuracy.

Our team invited a division 2 team to come visit this week. Our guests had been together for 5 years and have routinely won robot design and programming awards, so we wanted to pick their brains. I may write about some of the other things they shared, but I wanted to share one in particular with you.

Our robot was designed with the wheel base only slightly wider than the brick itself. The division 2 team told us we should widen it. The wider the wheel base, the more accurate the turns. They explained it like this:

When you program the robot to turn, generally you turn one wheel and not the other (by the way, they mentioned that one wheel turns are more accurate than two wheel pivot turns). A wheel base that is only slightly wider than the brick might need a duration of, say, 180 degrees to make the robot do a 90 degree turn. If you make the wheel base wider, it will require a longer duration to make the same turn. Let’s say a wheel base that is an inch beyond the brick width on each side needs 250 degrees to make the same 90 degree turn.

As I have mentioned before, programming a motor to turn a certain amount will have a certain amount of error. The motor may stop plus or minus a few degrees from the duration you specified. Let’s say the error is 3 degrees. Because the wider wheel base requires more degrees for a turn, the margin of error is less:

3 / 180 = 1.67%
3 / 250 = 1.20%

Therefore the wider wheel base will make more accurate turns.

While a 0.47% margin of error difference does not seem like much, you’ve decreased the margin of error by 28%. When you make a lot of turns, the margin of error will add up. The guest coach said generally after three turns, you’re lost. Of course, you will always use landmarks to reorient yourself, but minimizing the error will ensure you get as close to your mark as possible before looking for your point of reference.

Obviously, the width of your wheel base will be restricted by paths between the models on the mat. But you can find a happy medium between making the wheel base as wide as possible while still having room to maneuver around the models.

Our team widened our wheel base and we’ve calculated that we’ve theoretically reduced our margin of error by about 30%. I’m looking forward to seeing how it performs.


4 responses. Wanna say something?

  1. Dean Hystad
    Oct 27, 2010 at 14:19:27
    #1

    For the pivot type turns you describe, the amount of turn can be computed from the motor degrees using the following equation:

    Turn = motor degrees x wheel radius / track

    Track is the distance between the wheels. Ideally we sould be able to treat the wheels as if they had no width and contact the ground at a point. In practice the width of the wheel causes uncertainty in the track, and results in some turning error. I think difference between left and right turns in a robot is mostly caused by the track uncertainty.

    Looking at the equation it is easy to see how a change in track or a change in wheel size effects turning. Bigger wheels increase the effect of motor error, and a wider track decreases the effect of motor error. Of course motor error is only part (often not even the biggest part) of what causes turning to be off.

  2. Dean Hystad
    Oct 27, 2010 at 14:23:03
    #2

    For spin type turns where you spin the left and right wheels in opposite direction (steering in the move block hard over), the equation for turning is:

    Turn = motor degrees x wheel diameter / track

    Comparing this to pivot turns you should expect the turning error for spin turns to be about double that of pivot turns. In practice it is often much worse.

  3. Angie Wright
    Oct 27, 2010 at 15:41:41
    #3

    Thanks for a great tip!

  4. fllCoach
    Oct 28, 2010 at 09:21:51
    #4

    Thanx for posting these equations, Dean. I took my guest’s at their word that a one wheel turn (one wheel moving while the other is fixed) is better than a two wheel pivot turn (one wheel forward, one wheel backward). It really helps to see the reason why.

Post a Comment