Poll on winning teams using practice bots

[IndySam]

There is no doubt that our on field performance has increased significantly since we started building a practice robot three years ago.

We benefit from all the things that have already been stated but there are also intangibles that building a practice/prototype brings.

It compresses your build time so you can actually build two robots. This forces you to make decisions quicker and be much more organized. It also makes you raise more money which in itself brings other benefits.

I have yet to find a downside to it.

[BJC]

We built a basically identical un-painted robot this year as a practice robot.

This year we used the practice robot to:

-Practice match driving (2:30 runs)
-practice specific manuvers including double and triple balancing, jumping over the bump and downfielding balls, and of course aiming and shooting under heavy defense.
-improved targeting and shooting software
-developed both a center and side bridge auto that shot 4 balls.

We also make every change we ever plan on making to the comp. robot to the practice robot first to see how quickly we can do it, any problems that may arise, and if we actually like the changes we are thinking about making.

These are those changes from this year that I can remember off the top of my head:
-improved internal ball guides into the throat.
-sped up external collector.
-Changed front drive wheels for better turning proformance (twice)
-Changed gearing from a AM 1:4 to the 1:2.54 and modified final chain reduction from 24:12 to 22:12
-moved battery mount to front of robot to move CoG forward.
-Took antibackdrive pins out of window motor (for turret)
-Made many fang profiles (for getting over the bump)
-Shimmed out our bumpers 1/2 of an inch in the front to accomadate new fang profile.
-Moved our hood articulator to accomodate starting the turret backwards for auto.
-Moved a pot for the collector/bridge manipulator that was giving us trouble to a better location.
-put a ball sensor in the throat of the robot to sense balls immediatly when they went into the robot.
-many others.

It is also worth noting that a simulated field (or actual field) is very important to getting solid practice out of a second robot. Most of the powerhouse teams that build practice robots also have full fields to practice on.

Regards, Bryan

[IKE]

Originally Posted by BJC

...
It is also worth noting that a simulated field (or actual field) is very important to getting solid practice out of a second robot. Most of the powerhouse teams that build practice robots also have full fields to practice on...

I used to believe this was true, but after surveying this past weekend, I have found it to be not true.
Yes many of the top performing teams have a full field. Most have a 1/2 to 3/4 field.

[Siri]

We won both Philadelphia 2011 and Hatboro-Horsham 2012 & MAR Championship 2012 with a practice robot (and a roughly full field, though we could have shrunk some). It definitely helps a lot, including driving, programming (both hybrid period and autonomous assists in teleop) and mechanical improvements. I think the biggest thing it's made us do is build more quickly, though. We're still not fast enough, but it's getting there.

We always want to build the practice bot first, but in 2011 we got them done at the same time (seemed impressive then), and in 2012 we finished it around a week after bag & tag. Our goal for 2013 is to get it built first, at least in terms of each system.

[Adam Freeman]

I voted yes to the poll. We did had a practice robot that was very benificial to our programming team before and during the season. It was disassembled before we headed to St. Louis for spare parts.

We have built practice robots in 2005, 06, 09, 10, 11, and 12. Sometimes they are useful for driver practice, programming, etc...other times they are torn down for spare parts because they are not being used during the season.

One of the biggest benefits we get out of the practice bot is that we learn what works an doesn't work, in terms of design, manufacturing, and assembly as we build the first robot. This helps improve the quality of build between our practice robot and our competition robot.

I would suggest building a practice bot if have time and funding too and you are going to use it to improve before and during the season. Do not build one if it is going to effect the build quality or completion of your competition robot.

Competing in Michigan we get lots of driving experience in matches. We did not practice at all in 2009 or 2010. Actually the practice robot in 2009 was torn down before the season to fix the competition robot. In 2010 the practice robot was used for driver tryouts (before the grabber was installed), then we competed in week 1...so no more practice was conducted. In 2011 we practice a couple of times before the season, but never in the season.

This year the practice bot was identical and fully functional. We practiced a couple of times before the season. Once before Waterford. Once before Northville. And, then we practiced triple balancing before we left for St. Louis (so much for that.. ).

We have about a 1/4 field that we setup and teardown everytime we want to practice. It usually takes about 10min to setup.

This setup works for us...it may or may not work for your team. I judge if we need additional practice by comparing ourselves to the rest of the teams out there, if more is needed we do it, if not then we don't.

As my good friend JVN says, your mileage may vary...

[Robert103]

Originally Posted by BJC

It is also worth noting that a simulated field (or actual field) is very important to getting solid practice out of a second robot. Most of the powerhouse teams that build practice robots also have full fields to practice on.

103 constructs a full sized field every year. I can tell you from personal experience that the amount it helps cannot be understated.

[Bruceb]

Full field envy. We dont even have enough room to set up the hoop and the bridge in the same place.
But that is another issue.
Thanks for all the input.
Bruce

[Solidstate89]

Our team has never built ourselves a practice robot. We have neither the time or the resources to devote to such an endeavor as building an entire secondary robot. As much as it would help our programming team in being to continue working on a secondary bot while the primary one has already shipped, we just don't have the ability to do such. The best we do is have the programming team work on a robot from a previous year until the one we're building for the current season is complete enough for them to begin actively working on it.

This ws the first year I can remember where our programming team had more than the last 3 days before it had be shipped to actually program though. They had almost a full week!

When it comes to making a practice field, we usually build what's necessary. This year we had the baskets and a bridge built and just placed them in the gym for practice. We've never built an entire field before as far having all of the pieces, but what we do build are the essential devices for us to practice with. When it comes close to shipping time we also usually combine with 378 - The Circuit Stompers and their portion of the field to test autonomous modes and competition driving.

The idea that we used with The SUITS at FLR where they fed us balls during the Hybrid mode as we shot them was actually first devised at the Kenan Center with 378 a couple of days before shipping. I was quite ecstatic to see the idea come to light, even if it was with another team.

[trilogy2826]

Since 2010 Wave has completed an identical practice bot. I will echo all the words said before and add, I can't imagine not having a practice bot to, well, practice with. The human element of this game is the most underdeveloped an often over looked aspect of the game. Even the powerhouse teams could have the best machine in the world and with no practice, fail dismally.

In 2011, Wave had a phenomenal year with well designed, well executed, piece of junk, will never, ever do again octocanum drive. The ONLY reason we did well with a sub-par drive was that our drive team spent literally 312 collective man-hours finding a way to "make it work" before champs.

That being said, we are extremely lucky to have to sponsors we have.

Muza Metal Products builds absolutely every sheet metal part we ask for for 2.5 robots. Without them, we would have to scramble even more than we do

Triangle Manufacturing (the same company that provides KOP Lazy Susans) machines any custom parts we need and has EDM capabilities which is great for gear manufacture

CR Meyer provides half a hangar where they keep one of their planes. We use this for a full plywood field setup.

BTW, we had to work our butt off to get and keep such great sponsors and we are fortunate to have such great partners.

Originally Posted by Adam Freeman

And, then we practiced triple balancing before we left for St. Louis (so much for that.. ).

Dang Adam, rub it in... Why was it so easy on the practice field? (balls fists in frustration)

[Wayne TenBrink]

1918 has never built a practice bot. However, we do whatever we can to maximize practice/access time. We have a drivable chassis by week 3, so it starts there. We have a functionally complete robot by week 5. We attend a practice "opportunity" on weekend 6. Being in Michigan, we get at least 2 district events with 6 hours of "out of the bag" time prior to each. We usually go to a week 1 event, so there is only one week off between bag day and "out of bag" practice time. This year, we did 3 districts - every other weekend, so we were never more than 1 week without driving practice/upgrade time. Prior to MSC and CMP, we removed the shooter (built as a removable assembly for that reason) and were able to upgrade/test it by mounting in on a wooden frame and powering/controlling it with our test mule (spare cRio, etc. on a board). Going to the 3rd district event cost a lot less than building a full practice bot.

A duplicate practice bot would be nice, but there are other ways to get most of the benefits.

[jwfoss]

FRC2168 was #1 seed at both NYC and CT (though we were knocked out in the semis do to control system problems) and we do not build a second robot.

Our robot was completed at the end of week 5, and the drivers had a week of fighting over the robot with the programmers. We were then lucky enough to compete at the Suffield Shakedown Scrimmage and win.

Our drivers are trained each year with previous robots and a kitbot that we work on now and then. New "drills" are created to get the drivers used to driving in situations similar to the current game. We take driver training very seriously, and in the offseason practice as least once a week.

While we would like to get to the point where we can build two robots we have not yet reached the support and resources we believe are required.

[Alan Anderson]

Originally Posted by mandrews281

The main downside was our single electrical panel. It had all our Jaguars on it. We use CAN bus and direct connect many sensors to the Jaguars. So we had to have connectors for all the wiring (power and sensors). We had cronic problems with the encoder signals and at Championships we soldered around the connectors for two encoders. One of our off-season projects for electrical is to find better connectors for the sensors.

The TechnoKats Lunacy robot had an electrical/controls box that was completely separable from the rest of the robot. All of the sensor signals went through a single DB-25 connector. We occasionally used a long extension cable to connect the box to the robot during software development.