> I just got back from the Internation Aerial
> Robotics Competition at Epcot Center.
>
{snip]
>
> Between the cost of the Differential GPS,
> the RF modem to send the GPS correction
> information and the difficulty with tree
> cover Differential GPS currently seems
> incompatible with low cost mowbots :(
>
Monta,
Great info, thanks. Too bad, DGPS would be perfect. OK, then, here are
some things I've done some work on.
TI has a system (TIRIS) whereby small beacons can be powered temporarily
via beamed radiation, data stored and read, and then left behind. They
are designed to be cheap identification modules for product streams.
Currently they only work if within 1 meter of the reader. If we could get
these to be powered and read from more than one meter away (say 5 or 10
meters), then these could be planted in the ground, and powered/read by
the mower as it came near. Triangulating signals from two of these would
yield the mowers position, and grabbing a third would allow error
reduction. One of the nice things about triangulating is that the relative
angle can be measured quite accurately for small cost if you have a
processor to refine the data (like fitting a curve).
Another idea I had was for passive beacons made of semi radar-reflective
concentric plastic spheres. The idea is that 2 or 3 spheres nestle inside
one another, with the arrangement of sphere radii unique to each beacon.
Beaming out a single radar pulse, generates a reflected waveform which
would be dependent upon the radii arrangement of the spheres. E.g., if
one beacon is an 8" containing a 5" containing a 3" sphere, and another
is a 7" containing a 6" containing a 2" sphere, the reflected waveform
would be unique to each, and thus identifiable. Take those beacons and
put them where they won't be disturbed, like the attic of your house. Use
them to memorize the 'path' of the mower while you mow, and have the mower
repeat the path from then on.
I hate to use beacons, because of the additional infrastructure required.
But if it has to be a beacon, it must be passive and hidden. Running
lines to ugly beacons around the yard kills the whole project in my
opinion.
Any comments? I submitted the nested sphere beacon idea for a Small
Business Innovative Research government grant, and they said it was a nice
idea but please go away anyway :-)
* Michael L. Ross/C33 | Lockheed Martin Eng & Sci Svcs Co. ****
* Robotics Department | 2400 Nasa Rd. 1, Houston, TX 77058*****
* (281)333-7094; (281)333-6071 fax* #include <disclaimer.h>*
* Finger me for my public key. mross nospam at outland.jsc.nasa.gov ****