Sustainability - 2014 - COTS parts

[Qbot2640]

I'm not sure I really understand the focus of this thread. There has been so much talk lately about COTS parts, and I really can't see how these parts are making that significant of a difference. The real difference, from my perspective, is something like a "virtual prototyping" that is being done through videos...and whether the source of those videos is the Robot in Three Days project, other teams showing their accomplishments early in the build season, or just similar systems that clever teams find doing searches - the information is going to be out there.

Unless I missed the Andymark item "Frisbee Shooter Assembly" or the Vex "Pyramid Climber Mechanism" it still looks to me like teams have to cut and shape and drill and screw - there's no COTS short cut for that.

[Alan Anderson]

Originally Posted by Qbot2640

I'm not sure I really understand the focus of this thread. There has been so much talk lately about COTS parts, and I really can't see how these parts are making that significant of a difference.

How long have you been with FRC? Ten years ago, the list of off-the-shelf gearboxes was essentially DeWalt drills. Omniwheels were totally custom components. If you've only seen the program in action for a few years, you don't have a long enough view to understand how significant the existence of robotics-focused suppliers is for most teams.

Being a part of the team which was essentially the launching pad for AndyMark gives me a different perspective on things, to be sure.

[EricH]

Originally Posted by Alan Anderson

How long have you been with FRC? Ten years ago, the list of off-the-shelf gearboxes was essentially DeWalt drills. Omniwheels were totally custom components. If you've only seen the program in action for a few years, you don't have a long enough view to understand how significant the existence of robotics-focused suppliers is for most teams.

I remember in 2003 or 2004, someone at HQ decided to show how easy to assemble the kit drivetrain was. Bear in mind that the Kitbot did not debut until 2005. They had a small group of average Joes assemble it during Kickoff in another room. The problem was... All 3 or so of them had advanced degrees, and as I recall, they didn't actually get it driving. No joke.

In 2005, there were a lot fewer immobile robots, something about a pretty bulletproof transmission and an easy-to-assemble frame included in every KOP. AndyMark also started at about that time, so more teams were able to get shifters and omni wheels. (AndyMark has since grown, to say the least.)

[Richard Wallace]

Originally Posted by EricH

I remember in 2003 or 2004, someone at HQ decided to show how easy to assemble the kit drivetrain was. Bear in mind that the Kitbot did not debut until 2005. They had a small group of average Joes assemble it during Kickoff in another room. The problem was... All 3 or so of them had advanced degrees, and as I recall, they didn't actually get it driving. No joke.

Your memory is pretty good. That was the 2004 Kickoff. Those guys were not Average Joes.*

And although they didn't succeed in getting the kit drivetrain to move during the live video feed, that drivetrain was nevertheless a significant improvement over previous years. Some good ideas take more than one try to reach their goals.

-----------

^* 3620 didn't start until 2011.

[Justin Montois]

Originally Posted by Qbot2640

Unless I missed the Andymark item "Frisbee Shooter Assembly" or the Vex "Pyramid Climber Mechanism" it still looks to me like teams have to cut and shape and drill and screw - there's no COTS short cut for that.

I completely agree. Most of the COTS items available are for drivetrain components. Most teams can now at least drive. And sure, COTS components and community resources have helped on that front.

However, the main reason teams exist for a year or two and then get discouraged and fold, is they simply don't have the experience to tell themselves they need to limit themselves. It doesn't seem to matter how often we talk about 'building within your means' and 'have an appropriate understanding of your teams resources' it doesn't seem to get through.

Looking at 2013 for example, if you're a rookie team and you see the pyramid, you might have an understanding for it's difficulty but not nearly the understanding that a 5, 10 or 15 year team does. So they try it, and in most cases fail. They don't have the experience to recognize how difficult some tasks are.

It's like applying for a credit card and you get declined because you don't have a credit history. A vicious cycle.

A possible solution (there are many), is during the presentation of the game at kickoff, FIRST should specifically outline that some of the tasks in the game are meant for teams that are new or lacking resources. This might limit the amount of teams that attempt strategies that are way outside their teams ability....which seems to lead to discouragement....which may lead to the team folding.

[Alan Anderson]

Originally Posted by Justin Montois

Most of the COTS items available are for drivetrain components.

I'd call them powertrain components. Things like planetary gearboxes tend to be useful more for manipulators than for driving, for example.

Originally Posted by Justin Montois

However, the main reason teams exist for a year or two and then get discouraged and fold, is they simply don't have the experience to tell themselves they need to limit themselves.

Wasn't there a recent survey/study that suggested the principal reason for new teams folding after a year or two was lack of funding? I don't think discouragement is all that big a factor -- many of the people I talk to say that a "bad" year is more likely to encourage them to try harder the next time.

[Justin Montois]

Originally Posted by Alan Anderson

I'd call them powertrain components. Things like planetary gearboxes tend to be useful more for manipulators than for driving, for example.

You're right. That's why I said most components, Not all.

Originally Posted by Alan Anderson

Wasn't there a recent survey/study that suggested the principal reason for new teams folding after a year or two was lack of funding? I don't think discouragement is all that big a factor -- many of the people I talk to say that a "bad" year is more likely to encourage them to try harder the next time.

While I'm sure funding is an issue, the focus of this thread seems to be on the competitive aspect of FRC, not all of FRC in general.

I also completely disagree with the notion that there is somehow a lack of trying that's contributing to the downfall of some teams and somehow failure will encourage them to "try harder". Let's stop promoting failure as the means of motivation to getting teams to a higher level.

In most cases, these teams work very hard, they just choose something outside of their ability. A little assistance focusing these teams on appropriate strategies in the first few years is a good thing.

[Foster]

Originally Posted by Justin Montois

You're right. That's why I said most components, Not all.

While I'm sure funding is an issue, the focus of this thread seems to be on the competitive aspect of FRC, not all of FRC in general.

I also completely disagree with the notion that there is somehow a lack of trying that's contributing to the downfall of some teams and somehow failure will encourage them to "try harder". Let's stop promoting failure as the means of motivation to getting teams to a higher level.

In most cases, these teams work very hard, they just choose something outside of their ability. A little assistance focusing these teams on appropriate strategies in the first few years is a good thing.

I'm the original poster, so let me try to put some of my questions out a little clearer.

I know there is more to FRC than the robot, got that years ago. But a lot of new teams are focused on building a robot first, doing other community things later on.

And I understand that most of the COTS parts are drive train oriented, so some scoring method needs to be designed/built. I didn't mean to imply that there was a "Andy Mark Climber" or an "IFI Shooter", but I did say that with the better selection of COTs parts, scoring devices would be easier to build.

My theory was that COTS parts, would make it easier for teams to get over the initial hurdles of getting a robot pulled together. And if it's easier to build a more competitive robot (vs Dewalt days) will we start to see more teams become sustainable.

It sounds like from prior posts that my theory is flawed, the lack of constant funding (and/or not having a multiyear funding source in place) and the lack of high caliber mentors will still be the biggest failure points.

[thefro526]

Originally Posted by Foster

My theory was that COTS parts, would make it easier for teams to get over the initial hurdles of getting a robot pulled together. And if it's easier to build a more competitive robot (vs Dewalt days) will we start to see more teams become sustainable.

It sounds like from prior posts that my theory is flawed, the lack of constant funding (and/or not having a multiyear funding source in place) and the lack of high caliber mentors will still be the biggest failure points.

I think the theory is for a large part, true. If we consider that any increase in sustainability, whether it be one team per year, or one hundred teams per year, is an increase in sustainability, then it would remain true. COTS parts alone are not going to 'solve' all of the sustainability problems in FRC, but they do make it a heck of a lot easier to remain competitive (as in, can execute the game challenge) without overtaxing already limited resources. With limiting resources, human and financial being the root cause

As an example, I look at all of the teams this year, that were able to take the AM Kit Bot, and use it right out of the box. Sure, it's not the perfect drive train - but if you want something that isn't going to break (when properly assembled) and is easy to use at the lowest level (learning to drive) but also isn't all that limiting at the highest level (you can build a $@#$@#$@#$@# good robot on the Kit Bot) then there's little out there that will beat the C-Base when all things are taken into an account.

In the last two years, I've seen a handful of rookies pop up within MAR, most of which use the Kit Bot, or a drivetrain derived from it, and regardless of how well their mechanisms work, the fact that they can drive without struggling usually allows them to contribute to the alliances overall efforts, which gives them the feeling of being a part of whatever successes that alliance may have.

The same is true for a lot of 'also-ran' veteran teams, that essentially 're-rookie' every so many years due to the loss of student knowledge, or the loss of a mentor, or just from years of 'not caring about winning'. I've seen and been a part of veteran teams that built 'not so good' (being nice) robots using questionable custom drive trains and under planned manipulators for years and years. Those teams then switch to the c-base (either as given in the KOP, or modified similar to the KBOS method) and their performance seems to improve dramatically. Part of this is due to the reliability of the Kit-Bot/C-Base, but another part of it is that they've now wasted little, if any time on their drive train and can concentrate all of their limited resources on some sort of scoring device.

This sort of mentality applies to other basic robot mechanisms as well, but not to the same extent that it does a drivetrain, since no one (as of now) offers a complete solution to building an arm, roller, elevator, claw, etc - but most of the hard bits are either commercially available or well documented enough that a team can get relatively close without trying all too hard.

All of that being said, I do agree with a lot of the posts above saying that there are other issues that play into the sustainability discussion, especially those regarding funding, and mentor time/lack there of. The only real way to fix either of these seems to be a two-fold approach, on one side we need to emphasize to new or struggling teams the need and benefit to getting out into the community (world) and finding new resources - both financial and human - but also stress the importance of effectively utilizing what they've got.

IMO, from what I've seen, it seems like the easier of the two points to get across to someone is the need to 'have more', since most teams will acknowledge when their resources are lacking. Changing a teams methods or ideology regarding robot planning and construction seems to be something that is either really easy and well received or something that is basically like beating a dead horse. There are a lot of reasons behind this, some of which make more sense than others, but one of the primary roadblocks seems to be that a lot of teams don't really care about being 'competitive', or at least in the traditional sense. So many teams out there are happy to field a robot that makes an entire competition without breaking, or are used to making it into eliminations one year out every four or five, that they just don't see the need to chase that next echelon of performance. Fixing this might require some sort of 'FIRST-wide' commitment to being 'competitive' at every level, which can be/is interpreted as not necessarily being part of the message of FIRST. depending on how you define competitive

In any case, one of the many blessings of the District system is that it seems to do a much, much better job of 'lighting the fire' when it comes to engaging both rookie-ish and veteran teams. The combination of smaller events, and more of them, increases a struggling teams odds to have a 'good' competition experience, which I think most people can agree that a single 'good' or 'great' competition experience (i.e. winning an event with a powerhouse or something, or even just making elims) can be the single event that serves as a catalyst to years of improving performance both on the field, and off.

[Ben Martin]

I think a combination of things has contributed toward greater team success

1) 2013 had a game piece that was manufactured more consistently than 2012's was, thus making it easier for teams to shoot in a repeatable fashion, and thus it was easier to score points

2) More COTS components at cheaper prices at several "one-stop FRC shops" makes it easy for teams to find affordable components to build their robots

3) Besides RI3D, having more teams in FIRST means more testing videos uploaded to Youtube and other sites. Also, all the old videos from past games don't go away--the online resources for teams are expanding each year. Teams like 1114 have excellent resources about how to optimize your drive. By the end of week 2, a little searching of Chief Delphi could show that a balanced 8" pneumatic wheel or banebots wheels could be used to shoot discs. There were also videos showing passive 10-point hangs, and shortly after there were videos showing how a bucket could be used to index discs.

I like having examples available to make having a competitive robot accessible to all teams, and I think having COTS used (like RI3D did) in the examples makes it even easier for teams to recognize what they need to purchase to make the 'template' robot, without having everything necessarily laid out with instructions like the KOP drive is.

Is selling COTS subsystems outright, like a COTS shooter or COTS hanger, really the way we want to go, though? If we look at 2011, a COTS minibot? We already have COTS drives and COTS transmissions, which certainly have dramatically increased the performance level of many teams. A LOT of integration still certainly has to be done. At what point do the individual components become so few teams lose the 'fun' of designing and building it themselves? I'm not sure.

[Oblarg]

Originally Posted by Ben Martin

Is selling COTS subsystems outright, like a COTS shooter or COTS hanger, really the way we want to go, though? If we look at 2011, a COTS minibot? We already have COTS drives and COTS transmissions, which certainly have dramatically increased the performance level of many teams. A LOT of integration still certainly has to be done. At what point do the individual components become so few teams lose the 'fun' of designing and building it themselves? I'm not sure.

I think this is a non-issue; I'd expect that FIRST would not condone FRC suppliers selling what essentially constitutes ready-made entire robots, and could trivially put an end to it by simply not giving them any information on the game prior to kickoff. To this end, I doubt we'll ever see COTS mechanisms that deal with year-specific game tasks.

[Alan Anderson]

Originally Posted by Oblarg

I think this is a non-issue; I'd expect that FIRST would not condone FRC suppliers selling what essentially constitutes ready-made entire robots, and could trivially put an end to it by simply not giving them any information on the game prior to kickoff. To this end, I doubt we'll ever see COTS mechanisms that deal with year-specific game tasks.

The only manual I have handy is from 2011, but similar (if not identical) wording was present other years.

Originally Posted by Section 4.3 Robot Rules

...COTS items that provide a complete solution for a major ROBOT function (e.g. a complete manipulator assembly, pre-built pneumatics circuit, or full mobility system) that require no effort other than just bolting it on to the ROBOT are against the intent of the competition, and will not be permitted.

[thefro526]

Originally Posted by Alan Anderson

The only manual I have handy is from 2011, but similar (if not identical) wording was present other years.

Oddly enough, this was not present in the 2013 Manual, although I believe every year prior to 2013 or 2012, that exact rule was repeated.* As I looked through the archived version of the manual, I found the most relevant rule and the 'blue' explanation box below it, and it more or less allows a team to purchase a complete sub-assembly from a vendor regardless of function.

*I went back and checked the 2012 Manual, and it reads the same as 2013's, meaning that 2011 was the last year there was a rule preventing a supplier from selling complete mechanisms.

Originally Posted by 2013 Manual

4.1.3.5 R15
If a COTS item is part of a modular system that can be assembled in several possible configurations, then each
individual module must fit within the price constraints defined in R12.
If the modules are designed to assemble into a single configuration, and the assembly is functional in only that
configuration, then the total cost of the complete assembly including all modules must fit within the price constraints
defined in R12.

As Example 2 implies, a vendor could in theory sell an 'Arm Kit', that is pre-assembled for $399.99 and have it be completely legal within FRC. (As the 2013 rules are written)

The change from the previous verbiage may be due in part to conflicts with the KOP Chassis, since technically, according to some rules it would be illegal - although with it being supplied in the KOP, it's technically exempt from some/most rules.

Regardless, I doubt we'll see wide scale proliferation of 'ready' made mechanisms any time soon, due in part to the nature of FRC games. The fact that the game is unknown (at least at some level) to even major FRC suppliers would mean that they'd have to prototype, brainstorm, and produce any relevant mechanisms in a very, very short period of time, and then have the produced in an equally short period of time - or take a gamble on having a series of components designed and ready for production (or produced) by the time kick off comes around. The counter argument to this would be Vex's Chassis from 2013, since it was basically designed, built, and ready to be shipped within a week or so from kick-off, so it is "Possible" - but reasonable to do with mechanisms? Maybe.

If anything, we're long over due to see the wide scale production of 'build your own _______' kits geared towards FRC robots. 'Black Box' mechanisms like Ball Conveyors, Elevators, Telescoping Arms, Etc are well now well understood enough that someone with the resources could very easily manufacture a kit containing the 'tricky bits' of the mechanism with the end users supplying raw materials in the form of aluminum extrusion or similar. It's funny to see that there are actually a handful of teams that currently use this method internally and have refined their designs to the point where the only thing that ever seems to change is the size of the system, but not the construction method...

[Qbot2640]

Just in case anyone was thinking I advocated complete subsystems in an off the shelf package with my earlier post....I definitely do not. I was just trying to express that the nature of COTS items that is currently available has not changed the way my team designs or builds very much. We did buy some belt pulleys and some other hardware type items for 2013...but we have always used the kit drivetrain - and we have always constructed our specialized systems from our own designs.

I will concede that the current stock of available items has changed the activity greatly since the early years - it was my impression that this thread (and other similar current threads) were speaking more to very recent changes and/or comparison to seasons only a few years ago.

[DampRobot]

Originally Posted by thefro526

Regardless, I doubt we'll see wide scale proliferation of 'ready' made mechanisms any time soon, due in part to the nature of FRC games. The fact that the game is unknown (at least at some level) to even major FRC suppliers would mean that they'd have to prototype, brainstorm, and produce any relevant mechanisms in a very, very short period of time, and then have the produced in an equally short period of time - or take a gamble on having a series of components designed and ready for production (or produced) by the time kick off comes around. The counter argument to this would be Vex's Chassis from 2013, since it was basically designed, built, and ready to be shipped within a week or so from kick-off, so it is "Possible" - but reasonable to do with mechanisms? Maybe.

If anything, we're long over due to see the wide scale production of 'build your own _______' kits geared towards FRC robots. 'Black Box' mechanisms like Ball Conveyors, Elevators, Telescoping Arms, Etc are well now well understood enough that someone with the resources could very easily manufacture a kit containing the 'tricky bits' of the mechanism with the end users supplying raw materials in the form of aluminum extrusion or similar. It's funny to see that there are actually a handful of teams that currently use this method internally and have refined their designs to the point where the only thing that ever seems to change is the size of the system, but not the construction method...

I completely agree. It's not a particularly good business practice for companies to build "frisbee shooter" and "frisbee intake" systems for the number of logistic and technical problems mentioned in this thread. Even if they could see us "2014 manipulator I" or "2014 manipulator II," I doubt most teams would use them. They wouldn't have any reason to think that they would actually work, and using them would probably raise questions of morality.

On the other hand, someone is going to make a lot of money making components for 233 style arms, 254 style elevators, good arm components, and decent worm gearboxes. There's a lot of "secret sauce" that goes into a 233 telescoping arm say, as well as manufacturing technology, and it would make a lot of money as a great COTS solution. Teams will buy them in droves if they're available, well made and useful.