Brian Lloyd-6 wrote:
> That sounds pretty darned good.

Complicated, really, but software, and learning systems is what my real
hobby is, the hardware is mostly just a sensor platform. The goal is object
recognition, ultimately, but the first steps will just be seeing what the
limitations of the data are. Sonar is very imprecise compared to visual
systems, and it presents unique challenges, since each ping is a single
point return with a certain amount of fuzziness involved. The big question
will be whether or not the fuzziness can be overcome, but that's why it's a
hobby.

--
View this message in context: http://www.nabble.com/Quadrature-Encoder-tp13322047p15921475.html
Sent from the OOPic mailing list archive at Nabble.com.



(You need to be a member of oopic -- send a blank email to oopic-subscribe@yahoogroups.com )

On Mar 8, 2008, at 2:40 PM, Shaggy wrote:

> Sonar is very imprecise compared to visual
> systems,

True. I really liked the scanning-LASER technique that someone at MIT=20=20
developed (if I recall properly). It appeared to me to be potentially=20=20
very accurate.

> and it presents unique challenges, since each ping is a single
> point return with a certain amount of fuzziness involved.

That is because the return is not from a point but rather is from an=20=20
area whose energy decreases as you move off-axis.

40KHz has a wavelength of about 8mm. All the good information is in=20=20
the wavefront there is no interference ... yet. I think that if you=20=20
want to use short sound waves you are going to have to bring the raw=20=20
return in and process it better.

> The big question
> will be whether or not the fuzziness can be overcome, but that's why=20=20
> it's a
> hobby.

How much time and processing power do you want to throw at it? As I=20=20
said, the scanning-LASER where the plane of the receptor is different=20=20
from the plane of the scanner (there is probably a name for this=20=20
approach and I apologize for not knowing it) struck me as being an=20=20
elegant and simple approach to the problem. If people don't like the=20=20
scanner drawing red or green lines on them, use an IR LASER. Most=20=20
modern imaging devices happily work in the near IR.

--

Brian Lloyd Granite Bay Montessori
brian AT gbmontessori DOT com 9330 Sierra College Blvd.
+1.916.367.2131 (voice) Roseville, CA 95661, USA
http://www.gbmontessori.com

I fly because it releases my mind from the tyranny of petty things . . .
=97 Antoine de Saint-Exup=E9ry

PGP key ID: 12095C52A32A1B6C
PGP key fingerprint: 3B1D BA11 4913 3254 B6E0 CC09 1209 5C52 A32A 1B6C

=20

=20


(You need to be a member of oopic -- send a blank email to oopic-subscribe@yahoogroups.com )