Subject:
|
Re: Robolab, byte codes and assembler
|
Newsgroups:
|
lugnet.robotics.rcx.robolab
|
Date:
|
Thu, 27 Nov 2003 23:13:59 GMT
|
Viewed:
|
9812 times
|
| |
|
|
In lugnet.robotics.rcx.robolab, Claude Baumann wrote:
> OK
>
> You do something great, and all the world is looking at it, finding the finest
> detail one could have made much better. That's the power of internet-exchange.
> Don't take this personally. I only wanted to remind the great job they did at
> LEGO's when they created the RCX for kids and youngsters... and now engineers
> and other skilled people juge for 5 years - and still play around with the toy.
> Isn't that incredible?
>
> Best regards,
> Claude
It really is.
I do think somebody made a mistake specific to the rotation sensor, though. I
haven't decided yet whether it's the firmware, the sensor, or just the overall
mechanism. I can understand how it got past quality control - sometimes you
have to count over 50 rotations before it happens (sometimes it happens after
two).
I just want to help these kids, which requires that I use Robolab to accurately
move a robot a specific distance, and the count that the firmware is reporting
isn't correct. Consider this "the air filter Apollo 13 needs."
My attempt at looping and averaging double samples of raw sensor values seemed
to sample too slowly, because it missed many more ticks than the firmware did.
So, I thought I needed to get more low-level to get rid of any "baggage".
Can anyone tell me how I might go about having more control over the sampling,
for the purpose of rejecting or masking bad data, without introducing the kind
of fatal inefficiency my first attempt had? So far, I've had one reasonably
possible suggestion, which would take lots of work and _might_ be competition
legal: patch the firmware. I'm still not sure exactly how I'd go about it, but
I can look into it. I just hope for a less "invasive" way.
|
|
Message is in Reply To:
 | | Re: Robolab, byte codes and assembler
|
| OK You do something great, and all the world is looking at it, finding the finest detail one could have made much better. That's the power of internet-exchange. Don't take this personally. I only wanted to remind the great job they did at LEGO's (...) (21 years ago, 27-Nov-03, to lugnet.robotics.rcx.robolab)
|
24 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
