why sooo many bad robots in 07

[Karthik]

Instead of jumping on Shaun's choice of words, can we focus on the issue he's brought up? The FRC is much more exciting when there are more quality robots on the field. A drop in quality across the board is a serious issue. I believe there are a few factors.

- As pointed out, many teams have tried to do too much this year
- Not enough attention paid to manipulator design
- With the Banebot difficulties, teams had to spend more time on their drivetrains, especially at their initial event. (Installing new carrier plates)
- In past games we had full specifications for our task. Last year we knew there was a 4' foot ramp with a 30* incline. This year teams were forced to try and figure out what the specifications would be. This was a great challenge, and simulates a real world engineering decision. Unfortunately many teams made poor assumptions. (28" is wide enough for a ramp, we only need to climb 15*, etc.)

If you want to talk about why all robots are great and how "it's not about the robot", do it in a different thread. Let's leave this thread to discuss what's caused the drop in quality, and what we can do in the future to improve it.

[Cody Carey]

Who has the right to say that another team tried to do too much? It really isn't any of our business. What they decided to build their robot to do is their choice... and unless somebody else payed their registration fee, their choice alone.

We can't do anything to improve the overall quality of robots, at least not like this. That is up to each and every team that chooses to compete. The way to improve the quality would be to provide guides, and to help the teams who came up with not-so-quality robots.

Instead of saying "Teams are building pieces of junk, they need to do better.", we should be encouraging people. If we want to effect the target group of teams, we shouldn't talk about them in a thread which details how crappy their robots are.We should be making whitepapers on effective robot design; telling them what some effective strategies for our teams have been. Each team is different, and what works for us may not work for the next guy.

All I'm saying is that we are all going about this wrong. We shouldn't be talking as if we are better than the teams in question. While some of us may be able to build a higher quality of robot, the teams that need help will not listen if it is put across this way.

[Billfred]

Originally Posted by Karthik

- With the Banebot difficulties, teams had to spend more time on their drivetrains, especially at their initial event. (Installing new carrier plates)

The carrier plate issue is the gift that keeps on giving; all these pictures on CD combined with what I've seen at two regionals just makes me skittish about using them on drive. (However, I do give props to the BaneBots gearboxes for making it easier than years past to go omnidirectional. If someone were to bring to market a reliable one-CIM gearbox with a gear reduction similar to the BaneBots, I think they'd make a buck or two next season.)

One other thing that might factor into the equation: the rack itself. It is, arguably, the most complex structure FIRST has ever put on the carpet. More importantly, it's a pain in the butt to really replicate short of building the whole thing. (I've noodled a bit with 1293's three-spider-leg mockup, and I've noodled a bit with the field-spec practice rack at Chesapeake. You can feel a difference.) Many teams with limited budgets or manpower didn't build a full rack; I don't know of one within an hour of Columbia. If you don't know how the rack will react, you can't be fully prepared for the rack--and I'm thinking this element might have caught some arm teams by surprise.

Originally Posted by Cody Carey

Who has the right to say that another team tried to do too much? It really isn't any of our business. What they decided to build their robot to do is their choice... and unless somebody else payed their registration fee, their choice alone.

We can't do anything to improve the overall quality of robots, at least not like this. That is up to each and every team that chooses to compete. The way to improve the quality would be to provide guides, and to help the teams who came up with not-so-quality robots.

Instead of saying "Teams are building pieces of junk, they need to do better.", we should be encouraging people. If we want to effect the target group of teams, we shouldn't talk about them in a thread which details how crappy their robots are.We should be making whitepapers on effective robot design; telling them what some effective strategies for our teams have been. Each team is different, and what works for us may not work for the next guy.

All I'm saying is that we are all going about this wrong. We shouldn't be talking as if we are better than the teams in question. While some of us may be able to build a higher quality of robot, the teams that need help will not listen if it is put across this way.

I don't think anybody's trying to hand out bottles of Haterade around here--they're simply trying to get a better appreciation for why the quality, real or perceived, of robots this season is off compared to past seasons. If we start noticing a trend, perhaps we can then work to address the ones that are clobbering a lot of teams.

[wilsonmw04]

Originally Posted by Karthik

If you want to talk about why all robots are great and how "it's not about the robot", do it in a different thread. Let's leave this thread to discuss what's caused the drop in quality, and what we can do in the future to improve it.

First year here so i can't comment on the "drop in quality".
What we can do to improve the quality of 'bots in the future is much easier to deal with. SHARE what you know. Find the 2-3 rookie, or young, teams in your area and help them out. Share what you know about, drive trains, scoring devices and the rules of the game. Share your resources: mentors, programming and tools. This will help quality, nothing else.

[IndySam]

The enticement of pre-made wheels for exotic drives caused many teams with no experience with this kind of drive to give it a shot. Many discovered that it wasn’t as easy as it looked within a six week build time.

[Bharat Nain]

Originally Posted by wilsonmw04

First year here so i can't comment on the "drop in quality".
What we can do to improve the quality of 'bots in the future is much easier to deal with. SHARE what you know. Find the 2-3 rookie, or young, teams in your area and help them out. Share what you know about, drive trains, scoring devices and the rules of the game. Share your resources: mentors, programming and tools. This will help quality, nothing else.

You are right. Many veteran teams are willing to share their knowledge. Moreover, there are many resources on CD and FIRST websites.

However, this thread is discussing the drop in quality of robots overall through FIRST. There are not as many robots putting up tubes as tetras in the 2005 game.

Due to the variable nature of the tube and rack, I think it is harder for teams to adapt. The tube has been inflated out of FIRST specs at regionals, unintentionally. Some teams only designed their robots to work with the official specs. As we have seen through the years, if you design your robot to work with extreme precision of the game piece/object, you are going to be disappointed(most of the time, atleast). Another common trend is more than usual teams end up dropping their tubes because they are not gripped properly. And then, they try to put it on the rack and waste even more time because the tube will most likely drop on the floor anyway. Paul, I hope your white paper will address some of these concerns for teams interested.

The rack jerking around at times with no effort doesn't help either. It looks like many teams take a long time to get into position, and then if the rack moves, they have to reposition. Granted, the stingers are there for a reason and you can use them to your advantage to hold the spider leg in place. Many teams don't/cannot use it. As Billfred pointed out, it is also hard to replicate accurately. We have had to reprogram the heights of the spider legs at every regional - but that isn't so hard.

This is just my opinion. I had no intentions of offending anybody/team/robot.

[Ken Loyd]

This is a tough game! We spent a great deal more time brainstorming this year than in years past. Early on we thought we had all the angles covered but we used the old wood worker adage, "Measure twice...cut once" and reevaluated our design. It was a full week before we started even the smallest assemblies.

I like to think the planning paid off. With the help of Teams 60 and 1013, we won the Arizona Regional. We missed hanging the auto keeper only once in all our matches. We hung 7 ringers in one match. We lifted several robots 15" off the ground (we decided against a ramp.)

I think many teams under estimated this years game. They took it easy early and never caught up.

Ken

[Warren Boudreau]

In FIRST, there are no bad robots. Just robots that offer different levels of challenges.

Instead of KISS, many teams work to the concept of MICE.

Make It Complicated Everytime.

[Beth Sweet]

Once again I'd like to emphasize Karthik's change of the word "bad".

I think that this year's game brought on both the "bad" robots and extensive defense. Both of this year's game objectives were very difficult. Trying to make a manipulator capable of lifting a ringer is quite the engineering feat with lots of little issues (getting the arm to move, making the gripper move as intended etc.) as does trying to make ramps. Since the ramps cannot be rampy on the field, you have to design something that deploys during the match. I've seen a lot of robots that either get 1 ramp to deploy, both ramps get stuck halfway... etc.

It's a good thing that FIRST is raising its expectations for us, that it's trying to make our robots better and more technical knowledge is required. But personally, I prefer something like last year, where an offensive move (such as low goal scoring) can still be effective, and even teams with few or less skilled engineers can create an offensive robot that can compete. Last year, that's all our robot did was low goal, and we could still be effective.

[Donut]

Once again this year, numerous teams missed practice (and actual qualifying) matches. This was a problem last year as well (Check out this thread: http://www.chiefdelphi.com/forums/sh...ctice+matches), but I think it was slightly worse this year (practice faired about the same; but at AZ there were a significant amount of more missed qualifiers than in 2006).

I think part of the robots not functioning as designed can be blamed on the little practice time teams have; many don't finish their robot until right before ship (quite common, I know this is typical for us), but in the past two years many have been missing all their practice matches and even a few beginning matches. In this case you have no practice time until your first actual competition match; if you find a problem with your design at that point, the quick match schedule required for FIRST events means you won't get a chance to fix it until Friday night at best, after 3/4 of your qualifiers have been played. Going hand in hand with missing matches, this is the first year I have ever seen teams not passed for inspection when competition started; close calls I've seen, but never actual missed matches due to this.

I think the point that you need to get as much practice with your robot as you can needs to be emphasized a little greater in the future. I did not see any robots that were poorly held together, asking to be broken, I just saw a lot that had a gripper that didn't work quite as well as hoped, or ramps that weren't able to deploy right in all situations. A team cannot expect to semi-finish with the FedEx guy at the door, then use all of the practice matches as time to finish the robot, then use Friday morning as inspection time, and do well at the competition (I've seen it sort of happen once this year... but it's not something you can expect to work).

And why wasn't alot of this been a problem in the past? I don't remember many missed practices in 2005 or 2004 (they happened, but it would just be 3 on 4 matches, not 1 on 0), and I certainly don't remember any missed qualifiers, with the exception of a robot being completely manhandled one match and missing the next one for it. Inspection never seemed to be as big an issue either.

[flightofone]

I'm surprised so many teams can actually accomplish what they do from scratch. We totally miscalculated the amount of time we needed to put into the pre-season and build season.

Being a rookie team with no experienced drivers, builders, programmers, or mentors, we tried KISS. Luckily the game design allowed us to quickly discard trying to do anything with the game pieces, but still be interesting. Even with KISS, we only got one of our two major goals in the time allotted, the ramp. The drive train was a disaster.

This game is very complex and subtle, and the design decisions are hard to get right. Even experienced teams seem to have miscalculated the need for speedy, firm/angled game-piece control, and the need for positioning under heavy defensive pressure either with an extendible arm or a drive train with low gear that can push them into position.

There is a wealth of information on everything except manipulators on this web site, so we are looking forward to anything you can contribute to that. Next year maybe we can figure out how to arm ourselves.

[UCGL_Guy]

Do one thing well --- I wish we could listen to our own advice.
Two weeks into the season another mentor and 1 along with three students visited our new rookie team 2165 in Bartlesville Ok. and there on about 5 tables in front of us was the kit of parts. No design and not really knowing where to start. We offered direction and advice. They accepted it and in KC the made it to the finals. They consistently place 4 ringers on the lower rack. they also had a fairly decent drivetrain. What allowed them to do so well? Dedicated mentors - I'm not sure but they had around 5 or so retired engineers guiding and helping them along the way. The critical need in FIRST is Mentoring support. We are starting way too many teams without good support network in place.
BTW we were beaten out by our mentees!!!! (Is that a word?)

[billbo911]

Originally Posted by UCGL_Guy

Do one thing well --- I wish we could listen to our own advice.....
BTW we were beaten out by our mentees!!!! (Is that a word?)

What happened here is called a win-win situation. Mentoring a rookie team is awesome. Having them beat you is a testament to your mentoring ability. Either that or K.I.S.S. has been validated once again!!!!


I fully expect (hope) to see 2073 doing quite well at the Davis/Sacramento regional. Why? A rookie team that stuck to K.I.S.S. (I know, predictions predictions....)

[IndySam]

Originally Posted by Beth Sweet

I think that this year's game brought on both the "bad" robots and extensive defense. Both of this year's game objectives were very difficult. Trying to make a manipulator capable of lifting a ringer is quite the engineering feat with lots of little issues (getting the arm to move, making the gripper move as intended etc.) as does trying to make ramps. Since the ramps cannot be rampy on the field, you have to design something that deploys during the match. I've seen a lot of robots that either get 1 ramp to deploy, both ramps get stuck halfway... etc.

I was designing this crazy complicated ramp when the students came to me with a cardboard mock up of a simple two piece gravity deployed hinged ramp that would be easy to make and use. We build the whole thing in about two days. Simple, effective and it worked great.

KISS is the way to go.

[mtaman02]

I think FIRST should provide a crash course on drive train design class including but not limited to: proper mounting of motors, correct chains and sprokets to use, proper chain tensioners to use and mount.

This class would hopefully help all those so many teams with keeping their robot out on the field instead of their parts =). Just my .02.