Good Meeting Tonight

A few issues got resolved tonight at our NPO meeting. It really feels like starting the US STEM Foundation was a good idea. Decoupling STEM from a high school whose counselors and administration could care less may seem like a bad thing, however it is a short term temporary measure that will enable us to pull in extra support. We really need that kind of support to synergize the many fragmented initiatives that the county has undertaken in the last couple of years.

The cool thing is, no longer will one or two people be able to spear head random self-serving initiatives.

It is a LOT of work, but it will be worth it, I think.

To change the culture, we need to inspire kids. Yet we also have to remove the resistance to that change. Maybe this year Dean should assign homework for teams to submit essays outlining the red tape that prevents their local culture change.

"Get Over It" FTC Game Solution

Less than a week into the competition season, I believe I have a viable strategy for a successful robot. However, this is simply the first iteration of a full strategy, and will be used as an anecdotal study for how reverse engineering can apply to systems engineering.

These are functions, not forms. Know the difference!

Autonomous:

1. The robot shall traverse the field and get within alignment and range in order to acquire batons.
2. The robot shall then begin dispensing batons from the dispenser.

• For non-magnetic batons, the robot shall launch the baton back to its 'home' side of the field.
• For magnetic or doubler batons, the robot shall store the baton within the confines of the robot for further use.

Teleoperated:

1. The robot shall continue until it possesses 2 magnetic batons or until no more batons exist in dispensers:

• Dispense batons
• Store magnetic batons
• Launch non-magnetic batons to 'home' side

1. The robot shall traverse back to its home side.
2. The robot shall bypass the rolling goals, depositing the doubler baton into the low goal if the robot possesses it and no magnetic batons.
3. The robot shall corral as many batons on the floor on its home side into the floor goal as possible.
4. The robot shall push the rolling goals close to the same-colored balance bridge.

Endgame:

1. The robot shall lift at least one rolling goal off of the floor.
2. The robot shall then deposit all possessed batons into a lifted goal.
3. The robot shall then balance on the bridge.

Predicted Points:

• Autonomous: Traversal (10 pts) + 15 emptied batons (2 pts each) = 40 points
• Teleop 15 - 45 batons in low goal (1 pt each) = 15 - 45 points
• Endgame: 2 magnetic batons (25 each) + doubler (x2) baton in center chute of rolling goal: 100 points
• Balance of 3 elements on the bridge: 30 points

• Maximum score if the stars align: 215 points.
• More probable common score (1 magnetic, no doubler, 1 element on bridge, only 15 batons in lower): 90 points.

Not bad.

Card Thieves

A few weeks ago my girlfriend's credit card account started racking up bogus charges. $211 at a Wawa in BFE, Virginia. $110 at a Walmart in Wilmington, NC. Everything in total for her was ~$500 in about 24 hours. The bank's fraud alert system caught it at that point. A few phone calls later and now her accounts are ok and the bank is investigating it themselves.

Yesterday I found out my bank card was getting the same types of bogus charges: a $7 "test" charge and then 4 subsequent $105 charges at Walmart in Williamsburg, VA when I was actually on vacation in Bethany Beach, DE making legitimate charges. Fun! My g/f and I think they bought 4 $100 items to hawk on ebay.

Here's the key to why I'm not worried: She and I are cross referencing our statements and sending the findings to the relevant authorities. We're pretty sure the card was double-swiped somewhere, the number sold off, and a new card manufactured with a different name on the card attached to my account.

Our cross-references will help to find the first guy (or girl). Really, that person deserves a punch in the face. Some jail time wouldn't hurt; s/he probably wouldn't get a job working with money for a very long time either.

For the middleman who bought the number and sold the card: cut off his/her thumbs. Well, maybe just the thumb on the dominant hand. Next time s/he commits a crime, cut off the other one. Third time, both feet. Oh and let's be nice about it and use anesthetics. I only have these feelings towards the middleman because usually it's a coward who's the hardest to catch since s/he is never on camera or in public when doing the crime.

Finally, we're to the end user of the card. In my g/f's case, it appeared the person simply used the card for everyday spending: groceries, going out to eat at Olive Garden and Waffle House, gas for the car (or RV...); additionally it was very spread out which means it could be anything from a broke vagrant to gram and gramps on their retirement road trip. In my case, this thief was very obviously simply trying to rape the account as fast as possible: hard to catch in the moment, yet I'm sure Walmart has cameras. In either circumstance, I'd say simple jail time suffices. These people obviously have no motivation in life to make it on their own.

The suspect restaurant of origin is a Mexican restaurant that's less than a mile from my house; it's the only restaurant we frequent enough for both of us to have paid for a meal. It's a same too; I really liked their fajita quesidillas.

Thoughts on Simulation

There are 4-7 Excel tools available to design a drive train for FRC. On top of that, they're scattered, incomplete, or only show a portion of the overall picture. Successful teams seem to keep their best cards close to their chest, and it's impossible to wrangle a valid answer out of CD without the ubiquitous responses that say "why", or "WCD4Life!!".

Unrelated, but perhaps just as important to the stakeholders of FRC design, the 5th Gear Simulation is ... lacking ... in many respects. I have respect for the guys who do 5th Gear at LHM, yet they don't seem to understand their target audience within FRC. 5th Gear is more used as a LHM recruiting tool than an actual FRC simulation tool for FIRST, though 2010's game showed its utility pre-competition season (balls on the walls anyone?). The developers (who I work with and still have respect for within LHM) still miss the mark with their goals of taking the 5th Gear stuff to Xbox Live (etc). Why? I don't think those guys understand the volatility of such a market.

So two unrelated circumstances have led me to a single conclusion after reading this thread on CD: a simulation for FRC drive trains that allows drivers to drive their ACTUAL drive train around an obstacle course is an unrealized need within FRC. Shops like AM and IFI may even be able to use advanced iterations of such a simulation to test new tools without having to go through the whole process of creating tooling.

Hmm.

The sim would also need graphs that output functions -- distance vs. time, speed vs. time, distance vs. gear ratio while holding time constant, etc. That way adjustments can be made if anything is out of the ordinary (e.g. using 2 CIMs on a 140lb drive train geared for 14fps -- Atlanta '08 anyone??).

Even more, it should have a section where programmers could script in their control algorithms to run their code on their drive train, allowing drivers to (dis)approve of the controls setup in advance of the build season.

Oh yea, strategy guys should be able to design an obstacle course or other field element system for the drivers to practice on.

Dear God!

Finally, the simulation should have everything savable. Want to load up Team 001's uber drive train from 1992? Simply have them save a file and transfer it to you so you can load it into your sim.

Preferred languages involved: Java & xml. Potentially jgraph, jogl, and Eclipse extensions.

Meh.

For now I'll stick to making a Java drive train calculator that allows one to easily derive the results of slightly adjusting certain features. Power consumption per second, Theoretical & Actual speeds, Power consumed during (skid steer), Maximum torque for a pushing match, and maybe a few other simple statistical outputs. Oh yea, the ability to drag/drop motors, gearboxes, and wheels (all COTS) is a must too. Expect for early Fall '10.