Subject:
|
Re: localization (was re: GPS)
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Wed, 11 Jan 2006 13:55:30 GMT
|
Viewed:
|
1691 times
|
| |
|
|
In lugnet.robotics, dan miller <danbmil99@yahoo.com> wrote:
> Turns out it's not as simple as having two GPS units compare notes.
No, you do in general have to have them looking at the same satellites in the
constellation at the same time, and even then there was a good bit of number
manipulation to do. But the group I cave with did this with two laptops (to
record the data; one of them processed it later) hooked up to hand-held GPS
receivers, and the software was either cheap or free, I foget which. The
*downside* was we sat in a field in one location for a couple hours, but that
was because (a) selective availibility, which has since been turned off (yes?),
and (b) we had no way to get the same satellites. In other words, the first step
was to throw away some large fraction of the data when the two GPS units were
using different satellites. Neither of those is a big problem with the NXT. For
one thing, (a) may have been fixed (and if it's stil;l on, well... there is an
operational russian system, and the Europeans are putting up their own (better)
constellation as well), and (b) the two NXT will likely be fairly close to one
another (seeing the same satellites) and can communicate via BT to establish
that they are in sync to do a differential GPS calculation, with data being
easily transfered betwen them.
I'm not saying it's easy. But at least with the NXT, it might be possible. A
group of semi-literate cavers with two GPS units and a couple of laptops did it
more than 10 years ago. Surely we can match that?
> They tend to switch on and off randomly.
Really? I thought they were always transmitting. Which ones are in view, and
how selective avalibility is turned on and off (before?) was an issue... but
again, internally the GPS units know which satellites they are talking to.
> Localization, even in a known indoor space, is actually quite a difficult
> problem. Outside is much worse.
Yes, very. But with a unit with built-in encoders (dead-reckoning without
taking away one of my prescious input ports - I never liked the rotation sensors
just because they hugged a whole port) and the possibility of GPS input, it's at
least a problem with a lot more handles on it for the NXT generation.
--
Brian Davis
|
|
Message has 1 Reply:
 | | Re: localization (was re: GPS)
|
| I've always been partial to ultrasonic rangefinding. It's not too difficult indoors under controlled circumstances. The old Nintendo PowerGlove used this technique. For robot localization, you'd probably want a minimum of three non-moving marker (...) (19 years ago, 11-Jan-06, to lugnet.robotics)
|
Message is in Reply To:
| | localization (was re: GPS)
|
| On another list, we've spent some time researching differential GPS. Turns out it's not as simple as having two GPS units compare notes. You have to know which satellites you are talking to, and that's complicated. They tend to switch on and off (...) (19 years ago, 11-Jan-06, to lugnet.robotics)
|
12 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
|
|
|
|
