why does it say"IFI>"?
 

I have loaded the camera code from Kevin into our RC. But, the problem is when we try to get it to run, the terminal window only displays "IFI>". And then camera tilts down.

What does this mean?

Re: why does it say"IFI>"?
 

Which version of Kevin's code did you load?

Re: why does it say"IFI>"?
 

Does the program light stay orange?

-Kevin

Re: why does it say"IFI>"?
 

It probably does stay orange. I used to get that "IFI>" when I had a "partial" code error.

printf("IFI 2006 User Processor Initialized ...\r"); /* Optional - Print initialization message. */

^-- That's the line that didn't print completely..

Re: why does it say"IFI>"?
 

Don't you need the operator interface connected to run the robot controller.
I think its a safety feature in case of radio loss. Thats probably what it thinks...

Re: why does it say"IFI>"?
 

Yes, you do need the OI connected somehow, radio or tether. However, the program keeps running once it is started, despite radio failiure. The outputs are shut off, but your code will keep executing, and you will keep seeing serial output (if it is programmed to do so).

As for this:
printf("IFI 2006 User Processor Initialized ...\r"); /* Optional - Print initialization message. */

That line doesn't execute in Kevin's latest camera code. That entire section is commented out, if my memory serves me well. You will get an "IFI>" line back, and I believe that this is the "initialized" message from the camera code. I get it every time I turn on power to the RC, so I don't think it can be a corrupt message or an error. If the CMUcam is working, you will get "Searching..." messages and the contents of the "T" (tracking) packets back from the camera (blob size, confidence, etc.).

If the camera tilts down, there could be a lot of things wrong. First, I've noticed that the cam is sensitive to ground changes. If I discharge static to the chassis of the 'bot, the servos twitch, and stay there. It appears that the servos only move when the output changes, but they "drift" if there is no change in the output...I haven't found a way to counter this yet. Also, this might be due to the servos being plugged into the wrong outputs. Kevin's camera code uses the RC's PWM output 1 for pan, 2 for tilt (I think). Don't plug the servos into the CMUcam's servo outputs. They aren't used here. I'm not sure what else could be wrong, but if you do come across another problem, you can post back here and we'll try our hardest to help.

Good luck, and gosh that was a long post...
JBotAlan

Re: why does it say"IFI>"?
 

I did assume that's the line the window was printing, so I could easily be wrong about that. However, I saw the "IFI>" message many-a-times during the '05 build, and it would always happen when there was a code error of the red/orange/oscillating-light-of-death variety.

I guess I should just wait for the thread starter to comment.

Re: why does it say"IFI>"?
 

The "IFI>" prompt is displayed when the master code goes into "program" mode. You'll see it reliably and repeatably when you hold down the program button.

Re: why does it say"IFI>"?
 

I think that this may be my problem because i did recall coming across this line when looking at the code. I didn't change it because I thought thats the way it was suposed to be. So the next time I am at the shop I will be sure to give it a try.

thanks for the help.
-nick

Re: why does it say"IFI>"?
 

Actually I think it shows that every powerup, I seem to remember using that as an indicator the RC was on when I couldn't see its lights. Perhaps its just my memory being faulty, I will check tomorrow.

Re: why does it say"IFI>"?
 

I agree with Alan. It shows up when you press the program button.

Re: why does it say"IFI>"?
 

I can confirm that it displays on every powerup, but not when you press the reset button. My question is, is there any way to disable it? It confuses something I have connected to the Program port. :ahh:

Re: why does it say"IFI>"?
 

I doubt there is a way to shut off the IFI> prompt. What are you using, connected to the program port, that doesn't like that prompt?

BTW, did you steal my "good rep is always nice" line? :)

Re: why does it say"IFI>"?
 

It's a top-secret piece of custom circuitry that will be revealed at competition. :D
Whay I can say about it is that it uses autobaud detection; i.e. it expect the @ sign to be the first thing sent to it on bootup. When it receives the I instead, it calibrates to the wrong baudrate.

And yes, I did "borrow" your line. :)

Re: why does it say"IFI>"?
 

If you have control of the mystery device's software, the solution is obvious: don't do baud rate detection. If you can't fix it that way, the only simple thing I can think of to try is to add a delay circuit that prevents data from getting through at all until some time after power is applied.

(More complicated possibilities abound, such as putting another custom circuit in the line that sends the @ character before the RC has a chance to foul up the baud rate.)