NEW 2009 Control System Released

[adman]

Originally Posted by Daniel_LaFleur

I, too, have experiance with NI equipment. Believe me, their staff is well equipped to handle any level of technical difficulty. And with the amount of equipment that FIRST will be purchacing, a support agreement should be supplied. I trust FIRST to make sure on the support aspect of this deal.



The weight is nothing but another engineering challange. And the adaptor boards are nothing but breakout boards, a standard in manufacturing automation.



More than likely FIRST is going to use the FPGA to have the DIO mimic PWM outputs. They also talked about on-board image recognicion (sp?) and OCR capabilities. Most likely these will be on the FPGA. FIRST may also include the data transmission encryption keys on the FPGA so that they will be much harder to be spoofed, and can be changed by pit admin with just a download.



From what I've seen, the upconverter they use is high frequency and thus little noise generated.

I too am a bit concerned about the power loss and power consumption of the controller (as we haven't seen any data on that yet), but I will hold judgement on those until I see the datasheet for the unit.



Labview is as powerful as any other object orientated language. The fact that it's objects are depicted by pictographs instead of words does not change that. Labview can also import into it any API (created from C/C++/Pascal/VB/C# etc,etc,etc) ... the only issue is the base processor it is compiled for and the hardware specific calls.



More than likely it'll be a stripped down version of NI Vision and will be imbedded into the FPGA



[sarcasm]
Oh my gosh,I've been doing it wrong all this time. I've been using LabView to control motion control systems for a manufacturing enviroment for the last 5 years. How could I have been so wrong
[/sarcasm]



My assumption (yeah, I know what happends when you assume) is that we will be getting a stripped down version of this compiler, and that it will allow calls from C/C++ (but possibly not VB or Pascal)



The IFI controller was $400.
We do not yet know the price of a second cRio, so for now I cannot comment on the value (or not)



Sounds like a challange to me.



No problem about the size of your post. You have concerns (and valid ones at that ... I just disagree, but thats my opinion). CD is a place for these discussions.

NIs and FIRSTs motivations should be questioned. As teachers/advisors/mentors we are charged with questioning things. However, in this case, I believe (my opinion) that you are reading too much into logo sizes and corporate shenanigains (although I'll be the first to say I can be wrong).



The above is JM(NS)HO

Sorry about I don't know what the JM(NS)HO is but thanks for the
review. The reason I want FIRST to contact me is that I was prepared
to run all 1500 controls on my production line for at my cost to help first teams have less money to pay out instead of
more for the new system. I dont want my company name on the banner.

Open forums do good. I took a chance and spoke many of the concerns
that occured at Nationals. I didn't mean to offend anyone. As I said
I really like NI and I am a LabView programmer but also teach HIgh School
Engineering Classes when I am not designing at my Company.

[Doug Leppard]

It is always nice to have a new controller to play with. When I got started as a mentor with FIRST in 2003 it was the first year of autonomous. The processor was a pig (no pun intended) and it couldn't do encoders, we had to build our own pre-processor to process the encoders and pass data to the FIRST processor. 2004 changed all that.

I am glad FIRST is staying up with the times. But I have concerns mostly along the price range. The Best thing about 2008 is that you could use a vex to practice things and they were close enough and cheap.

No way as a mentor can I afford to go out and buy the new system. Big question how are we going to get these systems into the kid’s hands?

How much will the FLL and FTC kits cost?

As far as power, most teams can not even take advantage of the 2008 processor. Even at Atlanta there were many teams that could not do any autonomous and most teams only doing 1-2 lines. We helped many teams to do a basic autonomous mode.

I am sure this new processor will have software that will make it easier for inexperienced teams to do more. But also it has so much power that the power teams can do so much more and thus will leave the weaker teams even further behind.

Bottom line my main concern is the cost of the new system, with something cheaper FIRST could get more of them into the hands of more people to learn from one another.

Just some thoughts.

But I am excited, once we get it you know we will work hard to maximize it.

[Daniel_LaFleur]

Originally Posted by adman

Sorry about I don't know what the JM(NS)HO is but thanks for the
review. The reason I want FIRST to contact me is that I was prepared
to run all 1500 controls on my production line for at my cost to help first teams have less money to pay out instead of
more for the new system. I dont want my company name on the banner.

Open forums do good. I took a chance and spoke many of the concerns
that occured at Nationals. I didn't mean to offend anyone. As I said
I really like NI and I am a LabView programmer but also teach HIgh School
Engineering Classes when I am not designing at my Company.

I apologize if I came off kinda heavy handed (wasn't my intention). My intention was to show my background with NI equipment (mostly LabView programming with NI or MCC DAQ modules), which seemed counterpoint to your discussions and concerns.

oh and JM(NS)HO = Just my (not so) Humble opinion

I, too, have my concerns about the new controller. I'm concerned about the level of abstraction that LabView and VxWorks gives, and that the students may be missing out on protocol and hardware control stuff. Curently, with the IFI stuff, setting a PWM doesn't require the kids to learn about the timing of the 'on' pulse, about hardware delays, or any of that protocol ... and I'm concerned that we may be getting farther away from that understanding and utilizing systems and controls that students have no idea how they work.

My other concern is that teams will be getting 1 of these units and if it gets damaged, some teams may not be able to pay for a replacement. Depending on the cost of a replacement, teams might be required to skip a year or 2 if their controller becomes damaged. It's my hope that FIRST gets a decent enough discount for a single replacement (or spare) unit so that teams don't have to worry about this (I know they are working on it ... but it is a serious concern).

[qnetjoe]

Doug has a great point there are alot of teams even at nationals that can only do a few lines of code. Before everyone here just complains, just remember that there is a growing divide between FRC teams.

For example, I have had PAY to mentor FRC teams for the last 5 years. Some of these school can't even afford wire. Plus there are teachers that just run a team for the stipend. Compare that to one team that I know that get 1.3 MILLION dollars a year, over 50 mentors and have there own 100,000 sq foot workshop, equipped with 30+ CNC machines (They also do a lot more than FIRST). This divide is not just in the financial realm, but also in # and quality of mentors/teachers, and access to resources. We need to realize this and work hard at creating a community so all teams can fulfill the vision of FIRST.

The new controller helps is a lot of areas. I have seen some of the best FLL programmers just get intimidated my the nature of C and decide not to pursue programming at the FRC level. One of the biggest benefits of the new controller is a similar and familiar interface across all levels of FIRST.

FRC hardware was becoming rather clickish. the ifi stuff does not scale to handle a lot of the sensors and buses that are out there (CAN anyone?). The new controller give us so much more and above all it gives us the ability to freely and easily adapt this to other challenges outside of FIRST. We need to give NI and FIRST a break and let them develop the new controller. IFI had years to perfect the controller and it was still far from perfect.

[NCollins]

Originally Posted by writchie

I suspect the Cisco example showed access points spread apart by some tens of meters. This is enough to reduce the interference seen at the access points.

Cisco had their access points only ten feet from each other and sent all of the data from a single point.

Originally Posted by writchie

I have no idea what 5.47% crosstalk interference means.

It means that through extrapolation of the power curves of a wireless network the percentage above -30dB is 5.47% of the area with respect to a full power range.

Originally Posted by writchie

I would expect that the total combined throughput of the 6 networks you suggest would be well below that of a single network. Operating 6 robots on 1, 3, 5, 7, 9, 11 would probably not be too pretty, especially with synchronous traffic. When the robots are in close proximity, they will overwhelm the signals from the much more distant stations they are communicating with. The traffic would have to be extremely low for this to be workable.

Yes there is a drop in throughput but the robots only need 19.2 Kbps. Wireless can run up to 11Mbps. This means there can be almost a 99% drop and the robots would still work fine.

[yoyodyne]

Adman,

The points you make are appropriate and well measured. You can always pound a square peg into a round hole if you use a big enough hammer. The benefits this system brings are outweighed, by its cost, and size, weight, and power (SWaP). I see this abstraction as a step in the wrong direction in terms of teaching embedded system concepts and showing students that they have the technical capability and financial resources to build a micro-controller based subsystem. As an example, we had team members build an ATtiny2313 based project in the pre-season and as a result we were able to modify those to make a custom programmable Robocoach transmitter and multi-channel receiver. One of the good things about the IFI RC in my opinion was that teams were encouraged to build sensor co-processors and at least for larger teams, that was an excellent way to allow more students to be involved in both the HW and SW aspects of the control system. Somewhere in this thread the comment was made that a CAN bus interface was not possible with the old RC but I think they failed to see that a CAN to RC (maybe serial, or analog) converter board project would be a great learning opportunity.

I think that a team really needs three control systems; one for the competition bot, one for some type of test/integration platform/mobility base, and a third for control system development. The VeX controllers were a very good low cost option as extra development platforms for our software developers. The one extra cRIO is going to severely limit the time each developer has to unit test their code. Maybe the new system will be so abstracted and easy to use that little or no unit testing will be required (humor) but in all seriousness hard lessons learned about data types, protected sections, what the volatile keyword means etc. are likely to be lost.

From a teaching perspective what was really valuable about the FRC programming experience was that it was the first time most students worked with something besides FLL block diagram code or the school CS classes in JAVA, visual C/C++/C#.

I think that we will put even more emphasis on the use of micro-controller based co-processors as a way for students to learn how to develop powerful low cost, low power affordable solutions.

[dlavery]

Originally Posted by lukevanoort

For those worrying about security - I was told by an NI rep that FIRST recognized the issue and now has NASA helping them develop a method of securing communication.

I am not sure who from NI would have said that, but it is incorrect information.

[JesseK]

Originally Posted by dlavery

I am not sure who from NI would have said that, but it is incorrect information.

During the mentor Q&A, the FIRST rep said that the security issues would be outsourced to a private firm. He didn't mention the name of that firm.

[adman]

I think we are starting to get a consensus here.

* We all want our students to learn how things really work.
* We all see students become somewhat intimidated sometimes by C at first.
* We all have seen students overcome their fear of C
* We know LabView is a great way to teach control system basics but
can mask the background info that leads to true product design
in the embedded world if that is our cause
* We are all very nervous about how much these systems will cost
* We see the split in minimally backed teams and the teams that
may have these already on order from NI.
*Also considering we are adding more parts to the system, the analog
adaptor plugin and the expansion board and the power inverter the
robot now has potentially more things to trouble shoot to keep the
robot running.

Let's all make solid proposals of how we can solve these problems of
implementation.

[Chaos204]

Am I the only one who has yet to see any video demos or anything on the site? not that i am antsy or anything!?

[Adam Y.]

Originally Posted by yoyodyne

I see this abstraction as a step in the wrong direction in terms of teaching embedded system concepts and showing students that they have the technical capability and financial resources to build a micro-controller based subsystem.

That's not what FIRST is about. It may be how your car works in real life but it isn't the ideal way to introduce kids to programming because embedded systems needs a hefty understanding of electrical engineering.

We see the split in minimally backed teams and the teams that
may have these already on order from NI.

As yoyodyne clearly shows the split was exasperbated by the old system as well.

[yoyodyne]

Originally Posted by Adam Y.

That's not what FIRST is about. It may be how your car works in real life but it isn't the ideal way to introduce kids to programming because embedded systems needs a hefty understanding of electrical engineering.


As yoyodyne clearly shows the split was exasperbated by the old system as well.

We have a wide spectrum of programming abilities and experience as I would assume is the case for most teams. In some cases previous experience came from FLL and FTC and in others it has been from their own development activities including of all things Gamemaker which is a pretty good introduction to OO methodology. There has always been the option to program graphically but our experience was that the students that were interested in Kevin’s code were able to take that knowledge and develop their own microcontroller projects some for the robot and some for themselves. I don’t think the teams are going to get the VxWorks source and probably not the BSP source so the opportunity to learn at this level through the FRC RC has been taken away. You don’t need to understand anything about electrical engineering to read a micro-controller data sheet and learn from example how to service interrupts and use integer math so you don’t need a 32 bit processor to run PID loops for instance.


Is what you are saying that the split with the old system was mostly in capability while the split for the new system will be financial? From my perspective there will always be a capability split due to the difference in engineer/mentor availability. The teams that were relatively strong with the old system will also be relatively strong with the new system but now, only the well financed teams will be able to buy more than two? control systems.

[dcbrown]

Originally Posted by neutrino15

I can has link? All I could find was the WPI page and some stuff on NI's site.. Not 100MB, no enough to get started.. If all else fails, be cool and upload a package to FileDropper or something?

I can't upload the the documentation because it is copyrighted and is not publically available - at least in any way I could find. You can download your own documentation from Wind River by agreeing to the licensing terms via the 3.6 30-Day Evaluation kit. This is the top link in the list.

The above evaulation kit has the full doc set for VxWorks. Just remember, VxWorks is multiplatform targetted so this is the generic doc set. Some components such as full MMU support may not exist in the cRIO environment. But all in all, the documentation is a place for the software mentors to start.

I'd start with the following

Wind River General Purpose Platform, VxWorks Edition - Getting Started - general overview

And then move on to:

VxWorks Kernel Programmers Guide
VxWorks Application Programmers Guide
VxWorks Device Driver Developer's Guide
VxWorks Kernel API Reference

Wind River Workbench User's Guide
VxWorks Command-Line Tool User's Guide
Wind River Workbench Host Shell User's Guide
Wind River Host Shell API Reference

Wind River Host Utilities API Reference
Wind River System Viewer User's Guide

And a whole lot more...

[Jay Lundy]

Originally Posted by dcbrown

You can download your own documentation from Wind River by agreeing to the licensing terms via the 3.6 30-Day Evaluation kit. This is the top link in the list.

I downloaded that too and it should be a great introduction to vxworks and the Wind River Workbench, especially with the simulator.

I'm not sure if Wind River is creating a custom workbench for FIRST, but it would be great if they could get a full version of the workbench to teams before kickoff so we can familiarize ourselves with it. The Eclipse plugin for the workbench seems to be pretty extensive, so even though the code we write with the WPI libraries may be different, just being able to explore the IDE before kickoff would help a lot.

Like dcbrown said the evaluation comes with a lot of docs. There's also some info freely available at their site:
The main product page for the workbench is at http://www.windriver.com/products/workbench/
Here's a good introduction: http://www.windriver.com/products/pr...-Tech-Note.pdf

[Los Frijoles]

Although I will not have a chance to work on the system (I am graduating and moving about 1000 miles away from my team), I feel that this system will be a boon to FIRST and that it opens many doors. Here is what I see from my perspective:

Pros:
-Labview is easy to work with...easier for teams with no programming experienced people to operate
-Much higher processing capabilities
-Possibility for better, faster, and more advanced programs

Cons:
-Big
-Heavy
-May be a bit of a change for people who are already experienced with Microchip microcontrollers