MentorBuilt News: Mentor confident team will have 2019 swerve done before 2018 ends

Glenn Falls, Iowa - 52 year old Michael Lewingam met us in a small cafe near his team's workshop to discuss their off-season plans. Although already November, Mr. Lewingam stated his team "is going to design, fabricate, and test a swerve this off-season."

The team's goals are going well so far he says. "Well, the main thing they've got going so far is a PowerPoint of team 1323's swerve. It looks pretty good but I'm not sure I trust how much is 3D printed."

Despite build season approaching in a few weeks, Lewingam is confident that they will have a functional swerve ready for the 2019 season. "Yeah, we have a good programmer and a lot of machining resources at the shop. We just ordered an Omio router that we have to get set up and then we'll have the swerve machined in no time," the man said, "I'm not worried."

The optimism of this mentor is charming, and we sincerely wish his team well this season. Perhaps a swerve drive will give them the edge on the field they so desperately need. "We're a good team, ya know," he says, "but sometimes we just have bad luck with getting picked. We haven't played in eliminations in two seasons, but it's all political. You know how it goes."

Indeed, we do, Michael, indeed.

- MB News


  1. i d e a l l y
    you take the sine and cosine of the angle of each module relative to the field, multiply by distance traveled in the last cycle, bam
    b u t
    every wheel has its own drift/imperfections/magic and you end up with two wheels out in the stands and one behind your driver
    one approach to rectifying this is throwing out a measurement from a wheel that's an outlier ("yoooo guys I went 4 m in the last 20 ms"), if it exists
    calculate the estimated center of the robot from the position of each module and the robot's heading (I prefer to use the term "facing" in this context)
    average these center estimates
    use this center and the bot's facing to update the estimated positions of each module
    this way your wheels can't leave the bot :P
    how you determine outliers is a mystery left to the implementor
    also good luck figuring out why the bot has very little drift when heading downfield and to the right but to the left and toward the alliance wall you accumulate significant error
    regardless of the bot
    or the starting orientation
    the blue wheels reduce the problem compared to colsons but they don't fix it
    @carpet :screm:


Post a Comment