NEW 2009 Control System Released

[Team1590]

Today I contacted NI and the rep told me that the CompactRIO costs $3796.00, but thats buying it strait from them. He didn't know the FIRST price. The chassis cost the most for some reason.

[chuckstudios]

The fact that each robot runs as its own access point is intriguing to me for a simple reason: robot hacking. Yes, security of your robot will now become an issue. When Dean Kamen and company were on Curie field to demonstrate the new control board, my friend happened to be on his laptop. What did he find in the wireless networks? An SSID of "NItro", the name of the NI demo 'bot. He logged into it (unsecured!) but didn't do anything because he didn't know that he was actually logged into the robot. In hindsight, he could have made it actually dance to Soulja Boy.

Something that upset me a bit was that they said we wouldn't be able to modify the VHDL for the FPGAs inside the new controller. That means we have no real idea of what's going on inside it, and can't unload any "special" tasks to it. Then in the same breath, they say the libraries will be hosted on a Sourceforge-alike site. That's not hypocracy, I swear >_>!

I also hope the router can be changed to whatever kind we like. A La Fonera running DD-WRT would be amazing for size and weight reasons compared to the DLink they had on the demonstration units. Oh well, I guess we'll find out more closer to kickoff... only 8 months to go.

Originally Posted by Kyle Fenton

Does anyone know if this is mac and linux compatible. I know that there is a Labview IDE for mac os x, but is there a compiler for the new control board

According to the information session, you can develop on Linux (and Mac presumably) but not load code. I think an open source loader may be necessary...

[JesseK]

Originally Posted by Eldarion

Does anyone know anything more about the vision processing?

Framerate?
Resolution?
Color or black-and-white?
Color based tracking or higher level (geon-based, SIFT, etc)?
Does it use the FPGA?

Framerate will be determined by the update rate at which the new Driver's Station communicates with the bot, which will be limited by the wireless bandwidth.

Resolution will be determined by the packet size of said updates.

The camera is color.

It is color-based tracking, however they didn't say you couldn't have two and it looks like it will be easy to do some limited multi-color shape tracking given the speed of the processor.

I'm not sure if it uses the FPGA to process the color tracking.

[KTorak]

My only concern is readiness and ability to meet the deadlines of FIRST. From talking with the NI reps at the pits, they are not very far along and many of their answers were: "it's in the works," "its being prototyped," or "we aren't sure yet.". They said they only got the prototypes they showed today about a month ago. Also they said the controller should have a final weigth of about 2.2 lbs. In terms of IFI, it seems like they will slowly phase them out, whether or not that is FIRST's intention. The rep I spoke to said they are looking into developing their own speed controller to compete with the Victor 884 from IFI. I could see the being a big blow to IFI and their connection to FRC. The only other thing they would have at that point is stuff like sprokets and traction wheels. I guess we will have to wait and see how things go.

[galewind]

Originally Posted by Pat McCarthy

The youtube video on this page reminds me of the "Powerthirst" video.

Yeah I just showed a couple of my students powerthirst and then the FIRST video, and they remarked at how similar they were.

I said to another when sitting on Curie (this new control system is full of MENERGY!)

[Justin]

Hi Everyone,

Knowing the FIRST community as I do I'm afraid this post might ruffle some feathers. However in the whole of this conversation, or in an of the documents, I haven't seen any talk of the security employed on the WiFi network that team's robots will now rely on. Is there any encryption whether it be WPA or proprietary from NI? Please note that I am NOT advocating anything malicious during a match but I can think of ways that the WiFi/ethernet/IP(?) nature of the control system could be exploited. Now I recognize that those in FIRST practice gracious professionalism and I would not presume to suggest that anyone involved with FIRST would attempt an exploit. However as we are all reminded to often not everyone in the world lives by the FIRST creed. I think a little prudence would be the responsible course for FIRST and NI in this case.

Can anyone speak to this?

Thanks,

Justin

P.S. I see the post a few posts up now and that is disturbing. Clearly this will have to be addressed, with respects to my thread any ideas what shape such security might likely take?

[neutrino15]

According to the WPI PDF, we will be able to develop in Eclipse (C/C++) as well as Labview.

Q What IDE(s) will be available for use with the new controller/programming language?
A Both WindRiver Workbench (Eclipse) for C/C++ and LabVIEW.

Are there any example code snippets yet? I am dying to dive right in!

[sparrowkc]

Originally Posted by comphappy

I am 90% sure that it is a bin package, so no problem there long as your libs are up to date.

I am so happy right now. No more virtualbox, no more usb to serial convertor...

For Next year, I want to see the controllers themselves running on WindRivers Linux distro

[JesseK]

Originally Posted by Justin

Hi Everyone,

Knowing the FIRST community as I do I'm afraid this post might ruffle some feathers. However in the whole of this conversation, or in an of the documents, I haven't seen any talk of the security employed on the WiFi network that team's robots will now rely on. Is there any encryption whether it be WPA or proprietary from NI? Please note that I am NOT advocating anything malicious during a match but I can think of ways that the WiFi/ethernet/IP(?) nature of the control system could be exploited. Now I recognize that those in FIRST practice gracious professionalism and I would not presume to suggest that anyone involved with FIRST would attempt an exploit. However as we are all reminded to often not everyone in the world lives by the FIRST creed. I think a little prudence would be the responsible course for FIRST and NI in this case.

Can anyone speak to this?

Thanks,

Justin

P.S. I see the post a few posts up now and that is disturbing. Clearly this will have to be addressed, with respects to my thread any ideas what shape such security might likely take?

The FIRST rep that was at the mentor discussion today said that FIRST is outsourcing this issue in order to ensure they get an expert on the matter. They have to ensure security while also ensuring 24 teams (aka Atlanta setup) are able to be on the same network with equal bandwidth. Essentially, if they gave out how they were deciding to ensure security, someone somewhere would be that much closer to hacking into it. He also implied that FIRST would have detection methods and mechanisms in place to figure out the source of any tampering.

[lingomaniac88]

Well, with a webcam, there are many new possibilities, especially with optical character recognition. I don't know what the game is next year, but I can tell you this: 2009 will be one heck of a challenge for us programmers.

I wish our robot had the cRIO this year. If we did, it would've went around the track ABNORMALLY FAST! It would compete against Kenyans! It'd run as fast as Kenyans! It'd run so fast, people would think it's Kenyan! Then there'll be a tie and it'll be deported back to Kenya!!!

[CyberWolf_22]

Originally Posted by neutrino15

Are their any example code snippets yet? I am dying to dive right in!

The most information I have seen about programming in C\C++ for the new controller is on this page.

[Madison]

If we're not getting a new control system each season, I hope that we can expect lower registration fees in 2010 and beyond. I don't know how it's possible to justify charging teams the same fees while giving them fewer resources.

[artdutra04]

Originally Posted by MikeDubreuil

The rep I spoke to said they are looking into developing their own speed controller to compete with the Victor 884 from IFI. I could see the being a big blow to IFI and their connection to FRC. The only other thing they would have at that point is stuff like sprokets and traction wheels. I guess we will have to wait and see how things go.

That seems weird, as almost everyone I know or have talked to who are involved in other robot competitions (such as Battlebots, etc.) regard the IFI speed controllers as among the best in the business.

Edit; If anything, opening up the allowable speed controllers to several brands that meet certain criteria would be fine. But to purposely just move as far as possible away from a product with a successful track record in various robotics competitions seems like a dumb decision.

[Andy L]

Originally Posted by artdutra04

That seems weird, as almost everyone I know or have talked to who are involved in other robot competitions (such as Battlebots, etc.) regard the IFI speed controllers as among the best in the business.

Edit; If anything, opening up the allowable speed controllers to several brands that meet certain criteria would be fine. But to purposely just move as far as possible away from a product with a successful track record in various robotics competitions seems like a dumb decision.

I've been watching various robotics competition shows lately and it really surprises me because the victor does show up a lot more than I'd expect and it's weird to see something on TV and know that I use the same exact parts as them.

[eugenebrooks]

This really isn't a very hard problem to solve.

The RC and the OS can share a secret and use a checksum
for packet authentication.

The field control system can use public key methods, so that
a secret need not be globally shared for packet authentication.

Encryption is not required and it is best that the field control
system can see the traffic in any event.

In my view, it would be a nice if FIRST used sound methods
make sure that the communications for the field control system
is not spoofed. Putting the methods out for public review
is the best way to make sure that the chosen means is sound.

I will add that setting up the C and C++ environment as an
open source development environment is the cat's meow for
the FIRST community. A applaud this loudly!

Eugene

Originally Posted by Justin

Hi Everyone,

Knowing the FIRST community as I do I'm afraid this post might ruffle some feathers. However in the whole of this conversation, or in an of the documents, I haven't seen any talk of the security employed on the WiFi network that team's robots will now rely on. Is there any encryption whether it be WPA or proprietary from NI? Please note that I am NOT advocating anything malicious during a match but I can think of ways that the WiFi/ethernet/IP(?) nature of the control system could be exploited. Now I recognize that those in FIRST practice gracious professionalism and I would not presume to suggest that anyone involved with FIRST would attempt an exploit. However as we are all reminded to often not everyone in the world lives by the FIRST creed. I think a little prudence would be the responsible course for FIRST and NI in this case.

Can anyone speak to this?

Thanks,

Justin

P.S. I see the post a few posts up now and that is disturbing. Clearly this will have to be addressed, with respects to my thread any ideas what shape such security might likely take?