Nearly Officially Official: cRIO Control System To Be Reused

[Cory]

Originally Posted by Bob Steele

From what I understand if you purchased this control system on the free market you would be paying $8-10,000. NI is giving us the opportunity this coming year of purchasing a second unit for $1000 ...

Bob, do you have a link to where this is stated?

[Dave Flowerday]

Originally Posted by Bob Steele

AND we are teaching students to use INDUSTRY STANDARD hardware and software.

Are you saying that we weren't using industry standard hardware and software before? I can't believe that anyone would claim, with a straight face, that Labview is more of an industry standard than C.

[Bob Steele]

Originally Posted by Dave Flowerday

Are you saying that we weren't using industry standard hardware and software before? I can't believe that anyone would claim, with a straight face, that Labview is more of an industry standard than C.

I am not denying that using C is an industry standard. I do suggest that the hardware we were using was not something that a young engineer is going to run into in the form we were using it. Certainly the limitations of the IFI interface had to be worked around and one could say that this is a good exercise for any student. I don't know anyone that was using the IFI controller in a real application other than perhaps in a Battlebot or other simple robot.

I have seen Labview/cRIO used in a number of aerospace applications here in the Pac North West... I have not, to date, seen it used in a robotics application in person although I have seen it used in robotics applications on the web.

Even though, to this point, no one has been programming the cRIO with C or its variants... I have to believe that the ease we will be able to use C on this piece of hardware will be much greater than with the older microcontrollers.

Cory,
As far as the purchase of another unit. This is information I gleaned from a conversation with our NI rep here in Seattle. I believe it was also discussed in Atlanta. I don't know much more than that at this point. I would certainly suggest that teams squirrel away an extra grand for this purpose.

I certainly don't know whether this will continue into future years or not.
I know that NI and its partners are investing a great deal of money into FIRST and that our rep here in Seattle is very excited about working with FIRST.

[dlavery]

We need to be kind of careful here (and elsewhere) not to let the hyperbole overwhelm the discusion. When extreme statements are made that are unsupported by verifiable fact or sources, or even just completely wrong, all they do is weaken an otherwise cogent argument.

Originally Posted by Bob Steele

This controller is nothing like anything we have ever used in FIRST before ... or ANY other robotics competition has ever used.

Actually, the cRIO has been used in robotics competitions before, as well as in many other applications. For example, the cRIO was used as the central processing resource by some of the DARPA Grand Challenge teams - and the Grand Challenge certainly qualifies as a robotics competition. Most notably, the Virginia Tech team used the system in "Rocky," their entry in the 2005 competition, and "Victor Tango," their entry in the 2006 competition (which won third place and $500,000).

From what I understand if you purchased this control system on the free market you would be paying $8-10,000. NI is giving us the opportunity this coming year of purchasing a second unit for $1000 ...

Like Cory, I am not sure where these numbers are coming from - either the high or low numbers. They are inconsistent with all the other values I have seen or heard reported. Can you cite a documented source?

Who knows... it may even make autonomous TOO easy to do...

Somehow, I doubt that. Making autonomous easier will just mean that FIRST will have to increase the difficulty of the autonomous tasks to maintain the "balance of complexity."

Certainly the limitations of the IFI interface had to be worked around and one could say that this is a good exercise for any student. I don't know anyone that was using the IFI controller in a real application other than perhaps in a Battlebot or other simple robot.

Again, that would not be the case. For example, NASA has used the IFI hardware in the development of several prototype applications. A careful perusal of this Gigapan image will reveal one example.

The purpose here is not to throw stones at a particular series of messages. Rather, it is to make the point that unsupported extreme statements and/or overstatement of facts can be damaging to the discussion. They weaken the arguments we are trying to make, and decrease our ability to be effective communicators.

-dave


.

[Richard Wallace]

Originally Posted by dlavery

... NASA has used the IFI hardware in the development of several prototype applications. A careful perusal of this Gigapan image will reveal one example.

OK, Dave. That thing is just cool.

Care to enlighten us about its purpose? What is a Spidernaut, and what can it do? Clearly its control (or at least the control of two of its legs) is within the capability of an IFI RC.

Is that a wireless AP that I see mounted next to the wheel on the right?

[Elgin Clock]

Originally Posted by dlavery

Like Cory, I am not sure where these numbers are coming from - either the high or low numbers. They are inconsistent with all the other values I have seen or heard reported. Can you cite a documented source?

-dave


.

Since nobody else stepped up to answer that question or query yet, here is a link to a post of mine with a link to a price on NI's website which is a little more realistic than 8-10k or 1k the original poster Dave quoted had written that above response to.
http://www.chiefdelphi.com/forums/sh...3&postcount=34
As far as a discount for FIRST teams, I'm sorry, but I don't see anything official (re: on paper, or announced in an official FIRST newsletter) about that yet, so basically, that means as far as I'm concerned, it's not happening.
(Yet, if ever... - but one can always hope.)



Oh, btw Dave, where can I get my hands on a Gigapan camera? (I'm only half kidding. lol)

[dlavery]

Originally Posted by Richard

OK, Dave. That thing is just cool.

Care to enlighten us about its purpose? What is a Spidernaut, and what can it do? Clearly its control (or at least the control of two of its legs) is within the capability of an IFI RC.

Is that a wireless AP that I see mounted next to the wheel on the right?

Check the Spidernaut web site for some of the details on this project. Yeah, it is kind of cool.

-dave


.

[Greg Needel]

Originally Posted by Elgin Clock

Oh, btw Dave, where can I get my hands on a Gigapan camera? (I'm only half kidding. lol)

off topic, dave I would love the answer to this question. I signed up for the info newsletter on charmedlabs like a year ago and still haven't heard anything. I would suspect that they are not terribly expensive to make and I would buy one.

[AustinSchuh]

Originally Posted by Greg Needel

off topic, dave I would love the answer to this question. I signed up for the info newsletter on charmedlabs like a year ago and still haven't heard anything. I would suspect that they are not terribly expensive to make and I would buy one.

I was perusing the Charmed Labs website recently (looking for info on the new version of the Qwerk...) and I ran across an announcement that says more info is now available at http://www.gigapansystems.com/ The announcement is still on their homepage.

[Kevin Sevcik]

Originally Posted by Bob Steele

I have competed in BEST and those components are NOTHING like the cRIO.
I have seen a working cRIO dropped from a two story building and land and keep working. (as a demonstration). This controller is nothing like anything we have ever used in FIRST before ... or ANY other robotics competition has ever used. It is made to take abuse.

Bob,

I realize the cRIO isn't like the BEST components, but you must admit that BEST students and robots are much less capable of inflicting damage than FIRST students and robots. I'm primarily worried about what it's going to be subjected to in the pits when people are working on robots, and what it's going to be subjected to during installation and assembly, as most teams have had to worry about ESD damage to their electronics before. I'm also worried about the various bumpers and sidecars. I doubt they're being built to the same specifications, but it sounds like we're planning on reusing them as well. Finally, the cRIO isn't going to protect people from themselves. Miswiring, misinstalling, and mistreating the connectors are all going to occur and I don't think the lan port on the cRIO is rated for using a CAT6 ethernet cable as a leash for a 150lb runaway robot.

I'm not saying the IFI controller would fare any better, but veteran teams had a fair number of spares hanging about to help themselves and rookies, and replacement units were $450 or so, depending on what you broke. We still don't know the replacement cost of the cRIO or the various other components, but we do know that there will be far fewer spares hanging about than in years past.

[Tom Line]

Bob Steele Wrote: I am not denying that using C is an industry standard. I do suggest that the hardware we were using was not something that a young engineer is going to run into in the form we were using it.

I have seen Labview/cRIO used in a number of aerospace applications here in the Pac North West... I have not, to date, seen it used in a robotics application in person although I have seen it used in robotics applications on the web.

Cory,
As far as the purchase of another unit. This is information I gleaned from a conversation with our NI rep here in Seattle. I believe it was also discussed in Atlanta. I don't know much more than that at this point. I would certainly suggest that teams squirrel away an extra grand for this purpose.

First, I would like to point out that the old IFI system, as built, could be applied as an example to an enormous amount of the equipment in any one of the hundreds of manufacturing plants in the US. Embedded systems with very limited memory are extremely common on shop floors. As is C programming.

Second, my initial opposition to this change was that NI is not a standard in use anywhere. This is their way of trying to become one. The same way that MS subsidized their software in schools, and Apple subsidized their computers in schools. Get to kids early so that they'll continue use down the road.

Finally, I guess the point of this discussion is to let First and NI how we feel. I wonder if they have considered what is going to happen to a low budget team when their one and only C-RIO system fries in the first or second match of a regional. Having more money is always an advantage. Now a critical component of the robot is going to hinge on having extra money to buy a backup.

Perhaps we can make a suggestion that during the rookie year a team be charged an extra $1k or something, and get TWO c-rio's.

[EricH]

Seeing as this has turned into a discussion of spares, etc., I have a suggestion. This is merely a thought exercise; I highly doubt it will happen in practice.

Here is how distribution/spares could work. Every year, the new components for the year (if any) are sent out as soon as possible to fully-paid teams. Rookie teams, of course, get the full system.

Now, every year, there is a "window of opportunity" to get spares at a discount. Between two dates, teams can order extra cRIOs for less. The catch? Each team only gets x discounted cRIOs total. So once they reach that number, they can't get another at the lower cost, unless they can demonstrate that they no longer have one or more for some reason. (N.B. The trouble is figuring out a good way of enforcing the limit.)

Now, what about replacements? Provide proof that a cRIO broke and get a new one. Better yet, provide HOW it broke so that the NI engineers can further idiot-proof the system. (Intentional breaking does not count--full price.) Or, send the unit in for repair/replacement.

And, at any time, any FIRST team can get a new unit, but outside of a) the window and b) broken controllers, they must pay full price.

[Bob Steele]

Since obviously my comments are unsupported. I will not continue making them.

Return to this thread after we get to the competition and we will all see what happens. We will check on prices then too..

My numbers on costs came from an NI rep... and the high end included the entire control system and not just the cRIO. Many of these components are custom for FIRST.

I do have to worry when we look a gift horse in the mouth...

Sometimes I think that the influence the FIRST community has on the industrial control community in general is not quite what we think.

i would like FIRST to become a greater influence... but to think that NI is doing this in order to make a significant gain in the real market is a not realistic.

I am not going to say more... whether you like the change or not... NI is providing equipment because they also believe in FIRST. Not for advertising.

Believe it or not..

R

[waialua359]

Not to steer too off-tangent, but being able to reuse the cRIO is a benefit in terms of, if you buy a spare in 2009 and being able to use it in 2010 and beyond.
We talk so much about spares, yet under the old system, we were not allowed to use a 2006 or 2007 RC this year on our robot. Inspectors looked at the bar code info on the side and told us to remove the RC, borrow a 2008 one from the spare booth, and return it at the end of the competition. I thought that was petty since there was no competitive advantage to having one. Its not like we ran an older version of the software.

[artdutra04]

Originally Posted by Bob Steele

Sometimes I think that the influence the FIRST community has on the industrial control community in general is not quite what we think.

i would like FIRST to become a greater influence... but to think that NI is doing this in order to make a significant gain in the real market is a not realistic.

I am not going to say more... whether you like the change or not... NI is providing equipment because they also believe in FIRST. Not for advertising.

// These are just some casual observations...

I have a hard time believing NI would invest $10 Million dollars in a FRC-specific cRIO just from the goodness of their hearts.

Just as was pointed out above, NI is getting students used to NI hardware and software at a young age. As they get more familiar with it, they will be more likely to use it with future projects in college and beyond - in other words, creating a larger market share for NI.

While NI (and every other FIRST sponsor) may like the goals of FIRST, they all stand to gain financially in the long run from success in FIRST - whether through more available engineers or through larger market share of their components. And that's most likely the only way they can justify spending millions of dollars in donations to their shareholders.