Friday, January 22, 2010

Well Done, Supernova

Supernova had a fantastic showing in Arlington Heights last weekend. What we learn is worth much more than what we win, but this season the team was able to both grow as a team and also earn the following awards.
1st Place Robot Design Award
2nd Place Robot Performance Award
3rd Place Ambassador Award

They were judged Friday evening for project, teamwork, technical and for the ambassador's award. On Saturday, after their 2nd robot match, the team decided they should turn down the power on the arm when retrieving a grey loop - as it was throwing it instead of capturing it. That's when we learned that the laptop hard drive had crashed!

We hadn't yet learned that we would soon receive call-backs in both project and technical. This is a very good thing, but we needed a computer for project. Thanks to iLego for their generous offer of a laptop, but luckily Mrs. Naughton had a backup computer in her car.

Paul frantically started copying our robot programs and project materials from a jump drive and got the backup computer configured just in time to grab a quick lunch, immediately followed by the tech follow up, an impromptu chat with Scott Evans! and then the project call-back (for which he was using the newly configured laptop for the presentation - yikes!) ... and then his table run! I single out Paul because he held it together through a really daunting series of high-pressure events over the coarse of a couple of hours - although the chat with Scott was an eye-in-the-storm before the project call-back and 3rd table run.

Mr. Evans was the epitome of gracious professionalism, complementing the boys on their accomplishments and discussing their robots.

In this shot, they're discussing the team's first robot design of the season - a very cool pneumatically actuated 4x4 that was abandoned in favor of our low-profile "dually" robot. He commented that older teams like ours are often too emotionally attached to a design to make significant changes, sometimes unable to recognize basic flaws that limit ultimate performance. He complimented them on having the wisdom to see that this robot was not going to meet the requirements and that they could come up with a better design. I'm paraphrasing, but this is the gist of his observation.

He asked what they were planning for next season and the team mentioned that their coaches were suggesting a move to robo-lab. He encouraged this in that robo-lab has a higher ceiling of limitations than NXT-G and, more importantly, is more similar to what they would be likely to encounter out in the real world of programming.

There's more to report, but it's getting late. I'll close by saying we're extremely proud of the team's accomplishments this season. Yes, they've been recognized at tournament, but they've also learned a lot, grown as a team and have been, and continue to be, tireless ambassadors for FLL. They are currently preparing for 2 presentations to our local community, which I'll report on when the time comes.


Friday, January 15, 2010

Good Luck Supernova!

Good luck to Supernova at the IL State tournament this evening and tomorrow.  Your coaches and parents are very proud of the great work you've done this season, including extensive project research, helping other teams, introducing FLL to others, and showing some tremendous teamwork.

I'd like to share a little bit about how they've used teamwork to improve their robot's reliability.  The last 2 robot practice sessions were typical of how they've been working in small groups and handing off their results and issues to the next group through the use of their log book.  For full disclosure, the hand off in this example was verbal (BAM really had to leave by the time they figured it out!), but the log book has been their primary means of tracking robot errors and fixes for the last couple of months.

They had originally planned on last Jockey practice on Sunday, but 2 of the teams decided to do a little more practice.  BAM (Bridget Andrew Michael) did some minor reliability tweaks and were rewarded with consecutive 390, 390, 390, 400, 400 runs.  Their final run of the day (with Mr. Naughton watching!) was a 340.  The "falling arm" in mission 2 was back.  They figured out what was wrong (and why previous attempts to fix didn't work) but didn't have time to fix it.

Danny Patrick and Nate practiced the next day and did the fix.  The Stop and Wait my-block was disrupting a parallel path meant to hold the arm up.  Patrick suggested replacing it with a modified "Stop and Wait BC", which leaves the arm motor (A) undisturbed in the main path while the parallel path keeps it from falling.  This way, the navigation using B&C motors would run identically while still fixing the problem.  Danny did the changes.  They got a 400 on their last run and the arm was definitely holding during mission 2.

Great job!

Sunday, January 10, 2010

Reliability - Looper-Scooper or Looper-Lopper?

The team has had limited time since their late regional to make any major changes.  After Batavia, we had a brainstorming session to decide on what should be done to prepare for State based on our robot performance and the feedback forms.

So far, they've revised the script, re-shot and edited the "present ghost" video, brought to mind their teamwork skills, and have worked in small teams to incrementally improve robot reliability while each of the three 3-man teams have continued to practice.  At this stage, the failures are less spectacular (that's a very good thing!), but that makes it a challenge to say what caused missed points on the board, since a failure may not recur with any regularity.  I've been coaching them on the importance of careful observation and it really paid off today.

Example: Joey, Paul and Alex were practicing and found Mission 3's implement was often hammering the first tall loop.  Joey noticed that their "ramp" was not long enough, hitting the first loop end-on, missing it and throwing off the rest of the run.  They increased the length - but then it was out of base.  So they relocated it further back - but then the implement was lopping the first loop in half.  Everyone was watching it, run after run, until finally Alex saw that the top of the implement was now hitting before the ramp - it was happening so fast it was hard to see.  They shortened that part and it then started scooping loops every time.  Now, it works every time.

The 3 jockey teams have done an exceptional job of tracking changes and passing on information to the next team using their collaboration tools.  I'm really looking forward to observing their reliability in Arlington Heights this coming weekend.