Subject:
|
Re: an idea, can someone tell me if this is possible/been done before/etc?
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Tue, 2 Dec 2003 21:46:03 GMT
|
Viewed:
|
1642 times
|
| |
|
|
In lugnet.robotics, Jim Choate <ravage@einstein.ssz.com> wrote:
> On Tue, 2 Dec 2003, Kevin L. Clague wrote:
>
> > What if we assume that each output drives a motor,
>
> Bad assumption but it would work, the RCX is too limited on I/O to be
> doing that sort of stuff except in the case of simple rollers/walker.
> Since legs have symmetry it's silly not to take advantage of it here,
> you've got to use it to keep the thing in some sort of static/dynamic
> balance. If you're going to use differential/planetary/co-axial/pullies/etc.
> then -use- them [1].
Well I was restricting the discussion to modifying my Quad242. I think you are
talking about something else. I think you were proposing using all three inputs
and outputs to run the motors and control switches. I was proposing using less
of them. Where did we get off track?
Please explain what you mean by legs having symmetry and it is silly not to use
that here. I'm not often called silly with regard to my LEGO models.
It is my understanding that using LEGO, dynamic balance has only been acheived
by Steve Hassenplug's LegWay and models similar to it.
I've tried it with accelerometers, but have not had much luck there. I think
that the torque used to reposition the COG is noted in the accelerometer, which
causes more adjustments... which cause feedback and oscillation. I bought a
rate gyro, but have not gotten it interfaced to the RCX yet.
>
> Basic engineering efficiency. My approach is Occham's Razor and the
> concept of 'elegence'. If there is -anything- that isn't demonstrably a
> requirement for the goal(s) then it's 'outta here!'.
I understand. I am also an engineer. Notice that very few of my robots have
decorative parts.
I prefer to not put anything in in the first place, but half the time I do take
stuff out because I've figured a simpler way to do things.
>
> With regard to the discussion of walkers and leg blockage. Let's assume
> for a moment that instead of using switches or something we decide to do
> something mechanical. For example a triangular cam attached to the end of
> the leg segment. It would allow the end of the leg, assuming sufficiently
> low compliance [2], to move up over an object due to the rotational
> translation of forces the cam would provide. It takes the horizontal
> energy and transforms it into vertical energy.
>
> You can also move the cam farther up the leg, say a gear with a pin
> sticking out of it. It would normaly be positioned such that the pin was
> at the bottom of the rotation at all times during normal operation.
> However, when the leg stalled the extra torque would be sufficient to trip
> some sort of 'switch' (eg a rubber band providing counter tension) once
> the torque got over a specific amount. The rotation of the gear would move
> the pin to the top, thus providing a tad extra leg clearance for the bot.
> It would need a bang/bang stop at this point just to make sure nothing got
> too torqued out of shape if the leg is still blocked.
Blockage is not something I've studied at all.
>
> [1] It reminds me of an article SciAm many years ago about using pullies
> and springs to do basic addition/subtraction/and/or/not. Pretty
> powerful, don't know if it's universal as in UTM.
>
> -- --
>
> Open Forge, LLC 24/365 Onsite Support for PCs, Networks, & Game Consoles
> 512-695-4126 (Austin, Tx.) help@open-forge.com irc.open-forge.com
>
> Hangar 18 Open Source Distributed Computing Using Plan 9 & Linux
> 512-451-7087 http://open-forge.org/hangar18 irc.open-forge.org
>
> James Choate 512-451-7087 ravage@ssz.com jchoate@open-forge.com
|
|
Message has 1 Reply:
Message is in Reply To:
60 Messages in This Thread:
 
  
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
