[Noisebridge-discuss] Filling order for MA3 Miniature Absolute Magnetic Shaft (Rotary) Encoders

Corey McGuire coreyfro at coreyfro.com
Wed Apr 15 06:00:10 UTC 2009


Right.

So Christie and I are Orb Swarm nerds, and we have the great fortune to work
with Michael Prados and Niladri, who are working on the code necessary to
turn all these noisy data streams - GPS, Dead Reckoning, accelerometers, etc
- in to useful data.  This is something I wish to leverage for "Common, Off
The Shelf" Swarm kits or instructable swarming robot kits.  On top of this,
Christie and I are working on the project known to some as "Tangrams" but I
refer to as "Crystal Quorum"...at least until the quorum has selected a
name.  Swarm has GPS, CQ probably will not, but will sense peers and locate
in relation to them.

Dead Reckoning gives useful data about how far you've moved over a short
period of time.  Not only that, it's cheap and easy.

So easy, in fact, that these sensors I'm getting an order for would be
completely overkill for such an application.  We'd use something cheaper
because we really don't care about the position of the wheel, just that it's
moved and, roughly, how far.  These Encoders are for our unique steering
model for the Quorum.

Back to the discussion, what we want is position sensing (GPS), momentum
sensing (Accelerometers), direction sensing (compasses or GPS), and intent
(dead reckoning). With this information, we can safely detect GPS noise if
we aren't moving, detect slip if we are commanding ourselves to move but
acceleration isn't sensed, and get finer detail about our position than GPS
alone because we'll know which direction we've been traveling and for X
distance. This is following the model that we've used in Orb Swarm to a T.


On Tue, Apr 14, 2009 at 2:25 PM, Michael Shiloh <michaelshiloh1010 at gmail.com
> wrote:

> my understanding is that wheels slip (especially when steering) and so
> after awhile errors will accumilate. thus, it's usually necessary to correct
> this every so often with position via gps, beacons, video or other image
> recognition, accelerometers, compasses, etc.
>
> most robotic systems i've been aware of use multiple systems, because each
> one has its flaws and blind spots.
>
> Christie Dudley wrote:
>
>> Dead reckoning relies on knowing exactly how far you've traveled and in
>> what direction.  It's my understanding that this is the feedback mechanism
>> you want if you expect to have any degree of precision on which direction
>> you're pointed and so forth.
>>
>> I'm also thinking about looking into a completely different system and
>> figuring out relative location by a couple of different mechanism.
>>
>>
>> Christie
>>
>> ---
>> Fear is the mind-killer. Fear is the little-death that brings total
>> obliteration.
>> - Bene Gesserit Litany
>>
>>
>> On Tue, Apr 14, 2009 at 2:12 PM, Michael Shiloh <
>> michaelshiloh1010 at gmail.com <mailto:michaelshiloh1010 at gmail.com>> wrote:
>>
>>    hey corey! nice to see you here!
>>
>>    isn't dead reckoning flawed? or are you tolerant of these flaws?
>>
>>    Corey McGuire wrote:
>>
>>        Hey all, new to the list but been to NB a dozen times.
>>
>>        I am filling out an order for a project Christie and I are
>>        working on.  The application we have is, we need position
>>        sensing for steering and dead reckoning for robots.  There are a
>>        number of considerations and we've had a tough time finding one
>>        thing that does it all:
>>
>>           * Absolute encoding
>>           * High Resolution (positions per revolution)
>>           * Cycles (many as low as 20,000 revolutions!)
>>           * Size
>>           * Pin-out (many encoders require up to 8 digital IO each)
>>           * Price
>>
>>        But with the advice of the good Dr. Foote, we may have found
>>        what we need.  It's almost too perfect!
>>
>>        http://tr.im/abmag
>>
>>          1. It's got Absolute encoding
>>          2. It has 10bit or 12bit resolution.  Up to 4096 positions per
>>
>>             revolution! (We were going to settle for 16!)
>>          3. Cycles are in the millions, depending on the load
>>          4. .3" in diameter
>>          5. 3 pins
>>          6. As low as $36 for single orders
>>
>>
>>        Ok, so these are 5 times as expensive as the encoders we were
>>        going to use, but the cheap encoders meant we were either going
>>        to need serious hardware to read all the data, or we were going
>>        to need an Arduino per encoder.  They may still have been
>>        cheaper, but not when considering our labors.
>>
>>        There are two models I am interested in.  They are identical
>>        except that one uses bushings and one uses ball bearings.  The
>>        ball bearing model has a long life, can operate at higher speed,
>>        is less weight and may be smaller.  The price difference starts
>>        at $6.
>>
>>        If anyone else is interested in getting in on an
>>
>>        MA3-P10-125-N
>>
>>        Quantity        Price
>>        1       $36.00
>>        10      $31.54
>>        50      $27.78
>>        100     $23.89
>>        Over 100        Call for quote
>>
>>
>>        MA3-P10-125-B
>>
>>        Quantity        Price
>>        1       $41.80
>>        10      $37.34
>>        50      $33.58
>>        100     $29.69
>>        Over 100        Call for quote
>>
>>
>>        We are interested in getting a bulk order in, not just to lower
>>        price, but because we thought others might be interested, and
>>        the more interested people we have, the more brains we have to
>>        pick should there be problems.
>>
>>        What say ya'll?
>>
>>        Ciao!
>>
>>        --        Success is a tasty spirit distilled from bitter failure
>>        --Coreyfro
>>
>>
>>
>>  ------------------------------------------------------------------------
>>
>>        _______________________________________________
>>        Noisebridge-discuss mailing list
>>        Noisebridge-discuss at lists.noisebridge.net
>>        <mailto:Noisebridge-discuss at lists.noisebridge.net>
>>        https://www.noisebridge.net/mailman/listinfo/noisebridge-discuss
>>
>>    _______________________________________________
>>    Noisebridge-discuss mailing list
>>    Noisebridge-discuss at lists.noisebridge.net
>>    <mailto:Noisebridge-discuss at lists.noisebridge.net>
>>    https://www.noisebridge.net/mailman/listinfo/noisebridge-discuss
>>
>>
>>


-- 
Success is a tasty spirit distilled from bitter failure
--Coreyfro
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.noisebridge.net/pipermail/noisebridge-discuss/attachments/20090414/ed5ba077/attachment-0003.html>


More information about the Noisebridge-discuss mailing list