[FTC]: FTC Game Platform for 2009-2010

[Rick TYler]

Suggestions for changes to the robot/software platform for next year?

Here's one: go back to the Vex platform. www.vexrobotics.com and http://www.vexforum.com/wiki/index.php/VEXnet_Upgrade. The Vex platform certainly isn't perfect, but IFI has made huge strides in improving it every year.

Why does FIRST want to struggle making the LEGO/Pitsco platform full-featured and reliable when they already have a system like that available, that is more reliable and cheaper to boot? Just wondering...

If FIRST is locked into a partnership with LEGO, how about

1. Smaller, more reliable drive motors (the current motors are too powerful for the size of the field, and their failures are well known)
2. Tank treads
3. Rack and pinion gears
4. Chains and sprockets
5. Better shaft encoders
6. More reliable motor speed controllers
7. Longer axles
8. More chassis/stuctural material choices
9. Omniwheels
10. Some sort of linear sliding system

and

A reliable robot control system. I'm afraid Bluetooth will never be adequate to the task.

[dooey100]

Coming from a rookie team, I would be very dissapointed if FIRST switched back to VEX next year. Regardless of which platform is "better" the cost of having to buy another complete kit next year would be too much. The program will be expenxive either way, and I know that the only way I managed to convince my school to let us have a team was by showing them that the high cost only applied for the first year, and that there would be very little additional cost next year.

That said, our team coulld always implement our ideas, and although we sometimes had to make workarounds to accoplish what we wanted we never had to abandon an idea because we couldn't do it with the parts we had. Could it use some improvements? Definately. Was it so bad that it should be completely abandoned? Definately not.

[ttldomination]

I can very confidently say that FIRST will not switch back to VEX, and there are a number of reasons.

1. FIRST has realized that some teams have put in thousands of dollars into these kits, and now that they've made the switch, they have to roll with it.

2. FIRST has a partnership/contract with LEGO. This contract cannot simply be thrown out the window. FIRST will stick with the kit for a while, so there are no issues about that.

Furthermore, this year, the game rules specifically said not to use linear slide packs because they would be definitely useful. Rick, you asking for treads and more variety of gears is understandable, I want them too. But what if FIRST trying to lead us away from the comfort of VEX? What if FIRST wants us to stop relying on Tank Treads to get the job done or stop relying on part "x"?

I just think that this kit is really unique. Agreed that the bluetooth is sort of a pain, and I think that kit upgrades are sort of a given.

[Mushrooshi]

My teacher ordered 26 VEX kits.

Lol.

[Rick TYler]

My teacher ordered 26 VEX kits.

Lol.

Let's see, would that be VRC teams 3033A to 3033Z for next season? :)
If they don't raise prices, you could register all 26 teams for $726.

[JesseK]

We're sticking with FTC more than likely, though we'll probably have to scale back the # of teams. The only major change we've talked about is swapping to RobotC from Labview so we can better integrate sensors. We gave Labview a good run, but we're not satisfied with the Lego brick's inability to process our autonomous routine in that language.

I second the omni wheels and linear slide system. Perhaps a secondary supplier could be introduced to give us the more complicated items.

[ttldomination]

My teacher ordered 26 VEX kits.

Lol.

...That seems a little...extremely.

I mean even if you were doing VRC. You would only need like 2 kits to start out and then everything after that can be bought out individually....26 kits...

[Mushrooshi]

...That seems a little...extremely.

I mean even if you were doing VRC. You would only need like 2 kits to start out and then everything after that can be bought out individually....26 kits...

Actually, I think that is for her school curriculum.

*weeps in corner wishing I had that much vex*

[Foster]

My teacher ordered 26 VEX kits.

Well you know what my grandfather used to say, "You can't be too handsome, too rich or have too many robot parts."

I agree with Rick, it's a good start on 26 teams. And it isn't much more than the FRC KoP

[ttldomination]

School Curriculum makes more sense. :D.

[wilsonmw04]

We're sticking with FTC more than likely, though we'll probably have to scale back the # of teams. The only major change we've talked about is swapping to RobotC from Labview so we can better integrate sensors. We gave Labview a good run, but we're not satisfied with the Lego brick's inability to process our autonomous routine in that language.

I second the omni wheels and linear slide system. Perhaps a secondary supplier could be introduced to give us the more complicated items.

Could you elaborate on the problems you had with labview this year? We were able to use various senor applications with the brick last year without too much difficulty and after a bit of a jump start from the awesome gals from Twisted Bots we were running Auto like mad.

As for adding other specialized parts, I would be hesitant to add them to the kit. When you add complex parts to a kit like this, you become reliant on that part and set limitations on yourself. Take a look at the robots you saw this year. Did the kit limit the designs? I don't think so.

What I would add is more of the basic metal and gears at a reduced price. The cost of extra parts is the limiting factor for my team.

[Rick TYler]

What I would add is more of the basic metal and gears at a reduced price. The cost of extra parts is the limiting factor for my team.

One of my biggest concerns is about the cost of the system. Compared to Vex it has much less parts variety, much higher cost, and the parts failure rate is higher. After three years I have a pretty good grasp on the costs. A really competitive Vex robot costs us $600-700. Our two-and-half team FTC program (we built two competitive bots and one minimally functioning robot out of three kits and a bunch of extra parts) would have cost about $3,600 if we hadn't had the returning team discount. Starting from scratch, competitive FTC robots are $1,200-1,500 each. That's way too expensive. Burning out or breaking a dozen 12-volt drive motors didn't help much, either. The Vex components, on the other hand, have been pretty reliable. After three years and a collection of about 60 Vex motors, we still haven't had to replace one, although we do have two dead Vex servos.

I'm not factoring in team registration costs, by the way. This is just hardware. An FTC program using Vex parts would be a much more affordable way to deliver a mid-sized robot STEM program, even with the $275 FIRST registration fee. IMO. YMMV.

[JesseK]

Could you elaborate on the problems you had with labview this year? We were able to use various senor applications with the brick last year without too much difficulty and after a bit of a jump start from the awesome gals from Twisted Bots we were running Auto like mad.


Quirky encoder feedback, where it seemed the encoder would stop responding. At UVA I heard of teams using some loop to get around this (like a heartbeat check), but why add that overhead when there seems to be an underlying problem? We had 5 sensors integrated into autonomous -- 2 encoders, 2 bump switches, and a sonic range finder. Yet the robot kept either bumping into the wall, or stopping too soon or too late. The reason we believe it's the brick and not the code is that the results would change in between tests even when the code did not change. We will program Atlanta's autonomous in RobotC sometime in the next couple of weeks, so we'll know if it's just our lack of knowledge or if it really is the brick's inability to process everything.

I'm not factoring in team registration costs, by the way. This is just hardware. An FTC program using Vex parts would be a much more affordable way to deliver a mid-sized robot STEM program, even with the $275 FIRST registration fee. IMO. YMMV.

Aye, but first you have to believe in how your school system implements a STEM program. Around here, usually classroom-based STEM programs are put in gifted programs. To me, closing robotics off to a select few 'gifted' students is not something I believe in, so everything we do is based upon the broader scope of students in the county. That's just around this area though.

[dooey100]

One of my biggest complaints is about the cost of the system. Compared to Vex it has much less parts variety, much higher cost, and the parts failure rate is higher. After three years I have a pretty good grasp on the costs. A really competitive Vex robot loaded with omni wheels, sliding rails and all those other "limiting" parts costs us $600-700. Our two-and-half team FTC program (we built two competitive bots and one minimally functioning robot out of three kits and a bunch of extra parts) would have cost about $3,600 if we hadn't had the returning team discount. Starting from scratch, competitive FTC robots are $1,200-1,500 each. That's way too expensive. Burning out or breaking a dozen 12-volt drive motors didn't help much, either. After three years and a collection of about 60 Vex motors, we still haven't had to replace one, although we do have two dead Vex servos.

I'm going to have to disagree with this one. We used the base kit plus about 150 dollars worth of parts and had a very competitive robot, seeding third and placing second at our tournament.

We burnt out exactly one motor, then we got smarter and didn't burn any more. The only part that I would say is absolutely necessary to buy in addition to the base kit is an extra motor controller.

[lynca]

If FIRST is locked into a partnership with LEGO, how about

1. Smaller, more reliable drive motors (the current motors are too powerful for the size of the field, and their failures are well known)
2. Tank treads
3. Rack and pinion gears
4. Chains and sprockets
5. Better shaft encoders
6. More reliable motor speed controllers
7. Longer axles
8. More chassis/stuctural material choices
9. Omniwheels
10. Some sort of linear sliding system


These requests sound familiar to VEX's product line ... I wonder if FIRST can at least make a partnership to get VEX mechanical parts involved in the FTC challenge.

[smartkid]

Guys, comon... I attempted a 3-point-arm with this new kit. This is something we've done MANY times with Vex.

With Vex we used to be able to build arms that were very slender, but robust, light and fast!

FTC, just no... You cannot build anything smaller than the U pieces that are about 1.5" x 1.5."' With Vex our build scale is .5" x .5." And soon they'll have the even smaller Vex mini parts out that should allow us to really build smaller details.

Vex is up to speed and continuing to accelerate, Lego's new FTC kit went ten feet from the starting line then blew up. The engineers are still attempting to figure out what happened and pick up the mess.

ON THE COST ISSUE: It is not fair to say that "oh our FTC team just paid x amount for FTC kit so we have to keep it." Bull crap, ALL of the established teams before this year had to spend a great deal on Vex parts which they should still have. Furthermore, what you paid (that everyone is complaining about) was discounted. The real kit costs $400 more.

FIRST destroyed FTC and I am very disappointed. But on a lighter note, my team (which has an FRC team, two FTC teams and a VRC team) will be going to Dallas to compete with good old Vex, internationally.

-Cody

[ttldomination]

1. I still stand by the fact that we were all too comfortable with VEX so FIRST presented a challenge.

2. FIRST may seem like a downer at times but they're not heartless. They HAVE seen how much we've put into it and they won't change the platform. Agreed teams have poured thousands of dollars into VEX, heck my team did too, but FIRST was confident that the parts wouldn't go to waste with VRC going strong.

3. I do not believe that the Tetrix kit "destroyed" FTC. I strongly believe that it simply redefined it.

4. For the teams using LabView, I was wondering where you guys got your support material. I went to the first tech challenge website but I found that to be little or no help at all...

[Abra Cadabra IV]

1. I still stand by the fact that we were all too comfortable with VEX so FIRST presented a challenge.

2. FIRST may seem like a downer at times but they're not heartless. They HAVE seen how much we've put into it and they won't change the platform. Agreed teams have poured thousands of dollars into VEX, heck my team did too, but FIRST was confident that the parts wouldn't go to waste with VRC going strong.

3. I do not believe that the Tetrix kit "destroyed" FTC. I strongly believe that it simply redefined it.

4. For the teams using LabView, I was wondering where you guys got your support material. I went to the first tech challenge website but I found that to be little or no help at all...

I very much agree with all your points. Vex was making things too easy. I'm not saying that the new kit doesn't need better or more varied parts, I'm saying that there were too many easy solutions the old way. Case in point: tank tread conveyor belts. Designing a conveyor belt this year took lots more creativity and engineering than did last year, and I'm absolutely sure that FIRST did that on purpose.

Also something to keep in mind is that this is the kit's first year. FIRST is going to listen to our complaints and they're going to try to make things better. The contents and cost of the kits are not going to stay the same; the cost was one of the biggest complaints this year and I'm sure FIRST is going to try and reduce it rather than lose teams. It's a pity they couldn't get everything perfect on the first try, but who can?

I honestly wouldn't give up on FTC yet. Just give it some time.

[Rick TYler]

I very much agree with all your points. Vex was making things too easy. I'm not saying that the new kit doesn't need better or more varied parts, I'm saying that there were too many easy solutions the old way. Case in point: tank tread conveyor belts. Designing a conveyor belt this year took lots more creativity and engineering than did last year, and I'm absolutely sure that FIRST did that on purpose.

Carrying this argument to its logical conclusion, FRC is the worst of all because of its almost unlimited parts choices. You want tank treads? Just buy a set meant for a snow blower. You want a chain? Just buy one out of the McMaster catalog. Building an FRC robot is way too boring -- all the parts are out there just waiting to be bought.

I reject this argument. I prefer the kit to offer a selection of the sorts of parts available in the real mechanical and electrical worlds so that our students get to experience typical mechanisms scaled for the competition.

[Abra Cadabra IV]

...Yeah, maybe I didn't think that all the way through. I'm not sure if it's entirely fair to apply this argument to FRC (there's many other rules to take into account regarding parts there that don't exist in FTC, such as cost and weight) but I certainly see where my logic failed and I thank you for pointing that out before I made an even bigger idiot out of myself. :yikes:

I do agree that the kits need more part variety. I really miss rack and pinion gears, linear sliders, and sprockets. If I ever implied that we didn't need those, I am really really really sorry.

I think what I said at the end still applies though - this is the kit's first season. I wouldn't expect FIRST to get everything perfect on the first try. This season may not have been the greatest, but it's too early to start saying that FTC has been ruined - we really need to see what happens next season before we can even start to say that.

[Gdeaver]

Both platforms could stand better motor- servo solutions. I would like to see an integrated motor and drive control solution. I believe the Dynamixel AX-12+ Actuator would be a good addition to both platforms. See this link for specs
http://www.crustcrawler.com/motors/AX12/index.php?prod=63
Both controllers would have to be tweaked to allow the high speed serial link, but the benefit is that the motors can be packaged as a program object easily. At 44$ they are not cheap, but how much does a motor and a encoder cost and they still are not thermally protected and do not give torque feed back.

[ttldomination]

The FRC argument doesn't really apply because FIRST's main GOAL for FTC was to present a challenge where all teams have the same parts to choose from.

I'm not suggesting that rack and pinion gears and other commodities wouldn't be nice, but just that some of the other things that we are used to were getting a little redundant.

But it would still make sense and be mechanically nice to have some of the parts that the VEX kit has.

But, I think it's safe to say that everyone expects the FTC to grow from where it is now. :D.

[Mushrooshi]

I'd add cheaper plastic gears, maybe with a reinforced metal framework? Imagine... rickshaw wheels, but it is metal, not wooden, and has plastic filling the voids. Not to make the metal gears absolutely gone though. I'd rather have plastic drive gears and metal arm gears at the same time. Hell, making a decent drivetrain can cost the team in half their budget.

Piston sort of stuff would be interesting to see, and would add great potential. Perhaps hydrolic (sorry, can't spell it) pistons, for heavy stuff like arms, and maybe pneumatics, to toss stuff. It would be interesting.

Sets of chains and sprokets would be awesome aswell.

[JesseK]

Both platforms could stand better motor- servo solutions. I would like to see an integrated motor and drive control solution. I believe the Dynamixel AX-12+ Actuator would be a good addition to both platforms. See this link for specs
http://www.crustcrawler.com/motors/AX12/index.php?prod=63
Both controllers would have to be tweaked to allow the high speed serial link, but the benefit is that the motors can be packaged as a program object easily. At 44$ they are not cheap, but how much does a motor and a encoder cost and they still are not thermally protected and do not give torque feed back.

Good luck with getting that servo approved...that's 3x more powerful than FIRST allows in FRC:

http://www.trossenrobotics.com/dynamixel-ax-12-robot-actuator.aspx
http://www.usfirst.org/uploadedFiles/8%20-%20The%20Robot%20Rev%20K.pdf <R51-B>

However, that servo would be perfect for NURC (http://www.h2orobots.org/)... hmm.

[ttldomination]

I have no problem with chain and sprockets as long they are DURABLE. The ones in VEX kits were plastic and the chains would tear off if they came under too much pressure, which generally was around lifting an arm or something.

[alan4cast]

FTC, just no... You cannot build anything smaller than the U pieces that are about 1.5" x 1.5."'


Sorry, wrong! Our arm was based on a single angle piece. That one piece was used to support the entire structure of our puck grabbing assembly (three servos and two pieces of polycarbonate). That one piece is strong enough to lift the entire robot - it is 16mm on a side (.64").


ON THE COST ISSUE: ALL of the established teams before this year had to spend a great deal on Vex parts which they should still have.


And VEX also has their own competitions - in which all of those teams are welcome to compete.

- By the way, the people I've talked to all agree that the fact that you can still use the structural parts of the VEX equipment negates most of the "additional cost" issue. I will say that what our team found out (we are new, paid the full price for the kit - and then bought some VEX parts for certain things) was that the VEX parts were inferior in quality to the Tetrix parts. The Tetrix parts didn't need the same reinforcement as we saw in previous years because the stuff is simply better!


FIRST destroyed FTC and I am very disappointed.


FIRST changed FTC. Yes change can be painful, but in this case, change causes the requirements for innovation.

Alan

[Mushrooshi]

To be honest half of the robots at Atlanta were the exact same type of mechanism. I hope next year the challenge will allow no straight forward solution, so that we see millions of different kinds of robots.

[wilsonmw04]

To be honest half of the robots at Atlanta were the exact same type of mechanism. I hope next year the challenge will allow no straight forward solution, so that we see millions of different kinds of robots.

sorry, EVERY year there is a robot design that dominates FTC. That's what happens when folks have a chance to change the design after seeing competitions. I remember VEX conveyor belts dominating over the past two years.

[Abra Cadabra IV]

To be honest half of the robots at Atlanta were the exact same type of mechanism. I hope next year the challenge will allow no straight forward solution, so that we see millions of different kinds of robots.

Yeah, that tends to happen a lot in FTC, but it was way worse than usual this year. My guess is that it's because of the limited parts of the new kit, and also because of the design of the game.

Face-Off only had one real way of scoring: the center area, the only part of which that gave really good points was the triangle. In contrast, last year's game had three different ways of scoring: high posts, low posts, and goals, all of which had roughly the same scoring potential (well, not the goals unless you were really good at defense). Combine that with the ease of getting pucks out of the rack and the limited parts the new kits had, and hey presto there goes design diversity.

A better game and greater part selection will go a long way toward fixing this.

[Mushrooshi]

What would also be good would be a game where there are obstacles that can for the most part disable the robot completely for the rest of the game. It was rare that a good robot would be doomed by the ramp, and in addition, the wavy obstacles and the rolley things stopped only a few robots, if any. All they did was pose a threat to autonomous, and even so, it was easy to avoid it.

[ttldomination]

I think that this year, there were plenty of good ways to score, and I do agree that teams tend to ... not "innovate" when they see another design they like on youtube or whatnot.

1. the obvious dump out and dump into the center. Lots of robots had this as a CORE, but the robots with the variations made the difference.

2. The epic scissor lift. The scissor life was extremely beastly.

3. Treads. yes, believe it or not, Team "Raising Corn to the power 3" effectively used lego made treads to create an extremely belting system that worked extremely well.

But each year, FTC has one core design, and a few creative designs from teams who took the time to think up an exceptionally innovative solution.