Learning From Failure

Today my team worked on our minibot deployment system. We finally came up with a design that we thought would work and proceeded to build a full scale prototype. As you can see from the attached videos, it didn’t work out as well as we had hoped.

First attempt:

Second attempt:

Third attempt:

It was really a good day. after some tweaking we got something to work. well, almost work:


I hope you get a good laugh out of this. We sure did.

We are taking the reliable >> speed approach for the minibot. Give us another week and we’ll have something that i think will be very dependable.

Good job! We are using a somewhat similar method of deployment, where the minibot climbs off the robot, and it seems to be working well. What are using to stick to the pole, and what is your method of turning on the motors?

Thanks for the courage to give us all a chuckle. Keep tinkering, you will get it.

A word of caution: If you have been to or seen matches from competitions so far, you will note that many minibots end up on the ground during teleop when they are knocked off of their hostbot. If your minibot is free to climb out of its hanger, it is also free to be knocked out of its hanger. Make sure it is captured by some means so it cannot be knocked loose before you deploy.

I have two words for you: Ouch and d’oh!

Seriously, this is engineering. We try things, learn WHY it did not work, fix that, and take tiny (iterative) steps forward. Good work.

Our deploy system has the minibot sitting on a square peg. That keeps it attached to our deploy system. When minibot hits the pole, there’s only one way to go, up.

You can use two round poles instead. Just be careful that NO part of the minibot touches the hostbot (e.g., square pole) once the minibot gets to the deployment line.

our original minibot prototype started out by using a magnetic tool tray one of our mentors had in his workshop:

we took the magnets out of the try and used those to attach the bot to the poll. We plan on using a 4 poll switch (if I remember the name correctly) to turn our robot off and a KOP limit switch to turn it on.

During one of our last qualification matches we had finished tweaking our deployment system and minibot. We launched it and then the thing died like 90% of the way up the pole. Found out later one of our motors died so we did not have enough power to turn our shaft fully.

Programming wise it took me a bunch of attempts before ending up with the autonomous that is pretty consistent that we used at West Michigan. It only missed like 2 times out of like the 12 times we ran it due to the drive jumping causing the robot to turn enough to be misaligned.

yep, that is on my mind as well. We want something simple to keep that from happening. Maybe some sort of hard stop to keep in from falling off? That’s a problem for Monday.

From the looks of your deployment motion, you should be able to do as we did, and simply have a stationary finger or other feature which holds down the minibot and restricts it from coming out of its slot. Then during the deployment motion it simply moves out from under this stationary feature.

Our minibot is made out of some of the aluminum provided in the KoP, 2 wheels, a motor, the battery pack, a touch sensor from the KoP, and a single neodymium magnet. The touch sensor switch thingy is in front of the minibot, so that when the magnet attaches our bot to the pole, the trigger is pushed in, and electricity from the battery flows to the motor, powering the top wheel. It’s our first minibot, and we’re trying to make another one. Below is a picture of it.


I love this sentiment, and the wording :wink:

Seriously, good work, don’t be afraid to fail, and especially don’t be afraid to embrace it and laugh at it. Like 1676s minibot, the only way to go from here is up!


Well I think we have something that will work. Comments and constructive criticism is totally welcome.