IFI Control System (OI/RC) - The Future???

[Stuart]

hmm too bad would have loved a "program" button to be put on the OI board.

mabe . . 2 years from now

[Alan Anderson]

Originally Posted by Stuart

hmm too bad would have loved a "program" button to be put on the OI board.

mabe . . 2 years from now

You can add your own if you want to. Put a simple relay between a digital output and the program header, so your program can commit suicide and make room for its successor on demand.

(It's likely that you don't even need a relay, and a straight connection between the signal and the PROG pin will work, but I'm not suggesting that you do that.)

[KTorak]

Originally Posted by Andy Baker

That inspector needs their hard drive wiped and the common sense upgraded from version dingbat 1.0. This is a perfectly acceptable practice, and it has been done and approved for many years.

The only place on a robot that you can't use Velcro is for a traction device.

...sigh


Andy B.

We weren't in the mood to argue with this judge. He/She was kinda anal about everything detail...

[eugenebrooks]

Originally Posted by KTorak

At NATs, we were told we couldnt use velcro and had to remove it in favor of zip ties.

This was inconsisent with our experience. We velcro the backup
battery directly to the RC, using a patch of velcro that covers the
entire side of the battery, in a spot that leaves the lights on the
RC visible. It never comes off in an impact, and in fact is quite
hard to get off in order to swap the battery. There was no objection
to this attachment method at the nationals.

Eugene

[lemoneasy]

Just something I want to put out there, since now 32-bit microcontrollers and ARM controllers are so affordable, are we going to see a shift towards these controllers soon?

[lukevanoort]

Originally Posted by KTorak

We weren't in the mood to argue with this judge. He/She was kinda anal about everything detail...

I wonder if we had the same inspector on Galileo last year, s/he had problems with us wiring our fans to the inputs of the victors they were on. (As per IFI documentation) So, the robot had to have every fan (luckily only six) wired to our main power distro, and we had to scrounge up a bunch of 20A breakers. That robot now has such a terrible rats nest that I'm almost afraid to try and fix its wiring problems, due to fear of shocks.

[Astronouth7303]

Originally Posted by Jeff Waegelin

Or maybe just a more reliable radio system... I know my team and several others had problems with the radios at Great Lakes Regional this year, and there were also similar issues in 2005. Regardless of any other changes, I'm sure the new system will be more robust and reliable

We've had problems with radios in the past (I was sick this year, so I can't say).

I've heard that a different company (not IFI) designs the field controls, using IFI components. I don't know if Hatch does this or if it's some name we don't hear much.

[John Gutmann]

Originally Posted by Rickertsen2

I talked to The IFI guy at the Peachtree regional about the future of the controller. He told me that it has "outlived its useful lifespan" and that it was coming time for a new model. He told me that they had not yet decided what the new model would be like, only that there would be one.


I predict

• The PIC is here to stay
• Next year's controller will have a USB interface in addition to or in place of the serial port. I is likely that they will just add on an internal usb->uart chip.
• We may see a different connector scheme (I pray). It would require breaking compatibility with standard RC hardware, but would be well worth it. The can always sell adapter cables. the .1" headers are meant to be well protected. the do not take well to FIRST abuse, are not secure, and are generally a pain.
• A possible move from the traditional black flat square enclosure. This is WAY bigger than it needs to be. If you take it apart an look at the PCB, its very sparsely populated. Just look at the EDU.
• add an internal charger for the backup batt

I would also like but do not think it is likely

• An LCD display on the robot (2x16 characters min)
• Strong aux power on the OI

Is there a significant reason for predicting that the PIC is here to stay? THere are plenty other mC that are just as easy and even easier to use. AVRs, Gumstix, ect.....

The advantage of switching to another mC I think is that there is more support for the one they switch to.

Speaking from personal expirience and you can ask mike from 237 and matt krause I think it is about AVRs, but they are EXTREMELY EASY TO USE. You can get compilers that make libraries for you, I know of a few compilers that let you write code in other languages. So maybe for a rookie team they can program in PBASIC because they don't have much time to focus on that part. Then in off-season they switch to C

Also wouldn't they want to put the LCD on the OI rather than the RC beause then you can display diagnostics while in a match. And other messages, like you could have done shooter speed this year, or just plain old text that will let you chanegs something significant on your robot. I don't really know It just seems alot more useful on the OI.

[John Gutmann]

Originally Posted by Rickertsen2

I talked to The IFI guy at the Peachtree regional about the future of the controller. He told me that it has "outlived its useful lifespan" and that it was coming time for a new model. He told me that they had not yet decided what the new model would be like, only that there would be one.


I predict

• The PIC is here to stay
• Next year's controller will have a USB interface in addition to or in place of the serial port. I is likely that they will just add on an internal usb->uart chip.
• We may see a different connector scheme (I pray). It would require breaking compatibility with standard RC hardware, but would be well worth it. The can always sell adapter cables. the .1" headers are meant to be well protected. the do not take well to FIRST abuse, are not secure, and are generally a pain.
• A possible move from the traditional black flat square enclosure. This is WAY bigger than it needs to be. If you take it apart an look at the PCB, its very sparsely populated. Just look at the EDU.
• add an internal charger for the backup batt

I would also like but do not think it is likely

• An LCD display on the robot (2x16 characters min)
• Strong aux power on the OI

Is there a significant reason for predicting that the PIC is here to stay? THere are plenty other mC that are just as easy and even easier to use. AVRs, Gumstix, ect.....

The advantage of switching to another mC I think is that there is more support for the one they switch to.

Speaking from personal expirience and you can ask mike from 237 and matt krass I think it is about AVRs, but they are EXTREMELY EASY TO USE. You can get compilers that make libraries for you, I know of a few compilers that let you write code in other languages. So maybe for a rookie team they can program in PBASIC because they don't have much time to focus on that part. Then in off-season they switch to C

Also wouldn't they want to put the LCD on the OI rather than the RC beause then you can display diagnostics while in a match. And other messages, like you could have done shooter speed this year, or just plain old text that will let you chanegs something significant on your robot. I don't really know It just seems alot more useful on the OI.

Originally Posted by Matt Krass

No pressure on IFI eh Jeff?

I don't think we're going to see any major changes, however I would definitely like to see a few rollbacks. Maybe back to the original non-glitchtastic PIC with a memory module hanging off it?

Definitely want screw head power connectors again. I like the standardized PWM connectors too much to trade compatibility for comfort. Hot glue works fine for that.

I know I'm going against the grain here, but does anyone else feel like we need to make less changes, not more?

I feel like they kinda need to reassess the whole system and see what is absolutely necessary. After that they can start to add fancy bell and whistles.

Like as to the radio problem, what if they changed it to a blue tooth radio but made it a standard part just like the radios now. Just plug and play.

Also maybe IFI can make some ambidextrous joysticks rather then FIRST giving us 2 right handed ones. I just don't see how you can build a robot in six weeks but can't figure out how to make ambidextrous joysticks. I even remember one team using pens for the handles!

[Kevin Sevcik]

Originally Posted by sparksandtabs

Is there a significant reason for predicting that the PIC is here to stay? THere are plenty other mC that are just as easy and even easier to use. AVRs, Gumstix, ect.....

The advantage of switching to another mC I think is that there is more support for the one they switch to.

Speaking from personal expirience and you can ask mike from 237 and matt krass I think it is about AVRs, but they are EXTREMELY EASY TO USE. You can get compilers that make libraries for you, I know of a few compilers that let you write code in other languages. So maybe for a rookie team they can program in PBASIC because they don't have much time to focus on that part. Then in off-season they switch to C

I feel like they kinda need to reassess the whole system and see what is absolutely necessary. After that they can start to add fancy bell and whistles.

Like as to the radio problem, what if they changed it to a blue tooth radio but made it a standard part just like the radios now. Just plug and play.

Also maybe IFI can make some ambidextrous joysticks rather then FIRST giving us 2 right handed ones. I just don't see how you can build a robot in six weeks but can't figure out how to make ambidextrous joysticks. I even remember one team using pens for the handles!

Gumstix.... They have a lot of power and engineering built into them already, yet they'd need even more engineering to start approaching the capabilities of the current RC, I think. And I don't think IFI is in the business of reselling other people's controllers anyways. As far as the ATmega processor... it seems comparable to the PIC. 8-bit, A/D, etc. I I'll take your word for it on the compiliers and libraries and such, but any product from IFI is not likely to be an open architecture where you can do whatever you please. So I'm not certain there's an advantage there.

For joysticks, IFI isn't in the business of providing them or making them. You'll note that there isn't a single one on their site, so you're barking up the wrong tree.

Finally, bluetooth. *shudders* Okay. So Bluetooth works in the same frequency band as 802.11b+g wireless lans, and is meant for low power devices with an operating range of 30 ft. This means it's highly susceptible to being blocked by people, metal and generally just about anything. It's also likely to be interfered with by networks at universities, laptops, cordless phones, and microwaves. on top of this, if you're actually implementing Bluetooth, it has to accept and deal with attempts by other bluetooth devices to sync with it. So basically, I think it's a lot more trouble than it's worth. Meanwhile, IFI's radios are pretty darn robust and have good range and deal well with noise, etc.

[Gdeaver]

Instead of Bluetooth look at Zigbee. A high computing power Zigbee module and a few components could replace the whole operator interface and cost allot less. The master proc and separate radio modem could be replaced on the robot controller by a Zigbee module. I've used couple Zigbee parts at work and the results have been very good considering Zigbee is just getting going as a standard.

[Billfred]

Originally Posted by Kevin Sevcik

For joysticks, IFI isn't in the business of providing them or making them. You'll note that there isn't a single one on their site, so you're barking up the wrong tree.

IFI does sell (second section) the KOP joysticks--I ordered three for my kiwi-drive project. However, I agree with the principle of the argument; either use the KOP ones or buy/roll your own.