What is localization ?

"pogo"

I think I've heard the Northstar name before, and if I'm not mistaken, it is
as you guessed a positioning system that tries to emulate how GPS works, but
instead of your receiver to acquire signals from different satellites, it
acquires signal from emitters distributed accross your environment (on the
ceiling for example) on known positions. Be aware, I could be completely
wrong on this.

Cheers

Padu

Close, but...Northstar is named after the method seafarers used for
decades to find their way around the globe: find the North Star, and the
position of it and the other stars would tell you (within reason) where
you were.

The ER product uses a projector to cast an IR image onto the ceiling,
like a planetarium would project constellations. An infrared image
sensor looks up to the ceiling, and using the unique shape of the
projection indicates both the orientation and position within the room.

Obviously the system needs a ceiling, and a fairly unobstructed view of
the ceiling. The robot generally has a better view of the ceiling than
walls, so in theory Northstar is better than IR lighthouses positioned
around the room.

Similar methods have been devised -- or at least suggested -- by some
folks that use the distinct patterns found on most ceilings for
localizations. These systems need to be trained. With the right image
sensor and image system, it should be possible to detect distinct
variations in all but the smoothest, most perfect ceilings. Ceiling with
lights and light fixtures, water marks from rain, cracks, weird patterns
in the acoustic spray or panels, and other imperfections provide a
better rendering subject for the system.

-- Gordon

1)the robot explores your house  makes a map
2)you turn the robot off, pick him up, and move him to
the other side of the house
3)the robot gets turned on again, and told to localize.
4)the robot takes a few scans with his sonar sensors,
and compares the local map (what he just scanned)
with the global map (the map he's made with previous
exploration)

The spot on the global map that has the most similarities
to the currect sonar scans is picked as the "I am here"
spot on the map.


As the robot roams around, wheel slippage, encoder
errors add up over time creating errors in the robots'
"I am here" internal variable.

You can calculate the accumulated error, and make
the robot localize when the error passes a threshold.

imagine being at a house that you've never been
inside, and you have a blind fold.  When you first
enter the house, you get a quick peek of the layout.

you walk around feeling with your hands (so you don't
bump into anything) until you have no clue
where you are at.  It's time to localize.  Pull the blindfold
off for a second and take a look where you are at.

You have a map inside your head that you made the
first time you peeked.  Over time error accumulated
in your estimation of your position, and you had to
localize...  Compare the map inside your head to a
scan of the room.

You might be interested to read the following book .... [24MB or so
download, BTW] ...

Where am I?" -- Systems and Methods for Mobile Robot Positioning

http://www-personal.engin.umich.edu/~johannb/position.htm


- dan michaels
www.oricomtech.com
======================

Thanks! I downloaded it and it looks very interesting so far!
JCD

in mobile robotics, localization refers to the problem of, given a map
and sensor readings, determine where you are on the map.  similarly,
mapping refers to, given sensor readings and exact positions, build a
map.  neither of these problems are particularly difficult.  however,
neither of these problems are particularly useful either.  what *is*
useful is doing them both at the same time, ie, simultaneous
localization and mapping (or "slam" as it is generally called).  slam
is considerably harder than either localization or mapping and is the
topic of ongoing research.

chris

Great explanations from everyone! Many thanks!