Subject:
|
Re: rotation sensor problem (different one this time)
|
Newsgroups:
|
lugnet.robotics.rcx.legos
|
Date:
|
Thu, 27 Jul 2000 15:14:24 GMT
|
Viewed:
|
1755 times
|
| |
|
|
On Thu, 27 Jul 2000, Paolo Masetti wrote:
> > There is a /lot/ of code involved, so I won't post it all, but suffice to
> > say that the sensor is sampled, moves, stops, is sampled again, and then
> > several seconds worth of floating point calculations are done before the
> > loop is repeated. (Some printing to the LCD is done at various points here
> > as well.) At some point in one of those loops, the sensor stops responding
> > to new inputs. Turning the robot off and then on again, or just restarting
> > the program, seems to fix the problem just fine. I've also occasionally
> > noticed that it seems to be dropping counts (which was a problem solved a
> > long, long time ago.)
>
> All sensors stops responding?
Nope, just rotation. Well, I can't be sure, since there are no other
sensors, but since it is not permanent (i.e., I can still force it to
register, just at the wrong times) I doubt I have lost the IRQ.
> I think that you have lost, for some reasons, the ADC sampling IRQ...
> If you loose this one you have lost the kick to continue sampling...
>
> Some part of code inside legOS disable IRQs... I'm not enough expert
> in H8 to confirm that you can loose IRQ signals and to diagnose
> problem, but we can investigate...
>
> Do you work with a lot of signals? You could try to call directly the
> ds_handler() function when all sensors stops to see if everything
> restart...
No signals at all. I'll try it anyway, though, just to see what happens...
> > Does anyone have any idea what could "interrupt" the sensor, or cause it
> > to stop responding? This didn't seem to be happening just a week ago
> > (though it is possible that I might have missed it then) so I fear that it
> > is related to our new patches...
>
> Some test code need to be write to catch the bug, but we are in
> process to change the whole sensor sampling process...
Yeah. So if I can figure out a way to fix it in this program we shouldn't
worry about it too much. Still very frustrating...
Luis
-----------------------------------------------------------------------
"Summertime... and the living is easy...
fish are jumping and the cotton is high...
So hush, little baby, baby don't you cry."
-Ella
-----------------------------------------------------------------------
|
|
Message is in Reply To:
3 Messages in This Thread:
  
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
