Scott F. Touchton wrote:
>
> You could do acoustic range finding, but objects may prohibit this.
>
> How about placing 2  low power transmitters in the room.   Place a 4 antenna or
> more (the more the better, and the more fun you can have with math!) whip array
> on the robot.  Multiplex the array into an FM receiver.  When you "scan" through
> the antennas you'll get a doppler shift as you "switch" towards and away from a
> transmitter.  This will give you a vector with a little number crunching... 2
> xmtrs give you 2 vectors (or 2 receivers if you do it the other way).  2 vectors
> give you your position in the room.


Another common way to handle this is to place
IR "beacons" up on the walls in various places about
the rooms. Uni robotics guys usually do this system.
Then all you need on the mobile robot is a rotating
sharp IR sensor module (like from a TV or $5 from
many suppliers), and a small stepper motor to rotate
it.

This has a lot of advantages as the robot can compute
position from triangulating the beacons very easily.
Only full sunlight causes problems, most other problems
like obstacles are no big deal as the IR tends to go
around them and the bot can just work with the "fuzzy"
vector for now until it's next move.

A laser rengefinder is nice, but a commercial unit
like the Sick brand costs many thousands of $$ and
home made units are a lot of work. Unless someone knows
a cheap source??
-Roman

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads