Subject:
|
RE: Autonomous Robot
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Thu, 10 Aug 2000 13:00:17 GMT
|
Original-From:
|
Sattler Chris-QA1406 <CHRISTOPHER.SATTLER@stopspamMOTOROLA.COM>
|
Viewed:
|
418 times
|
| |
|
|
Would a cylindrical mirror work as well as a retro-reflector? Once you hit
it dead-on it would reflect back. It would reflect the laser beam to other
parts of the room, but they would be limited to the horizontal plain in
which the laser is already sweeping. The only problem would be if the
cylinder was not vertical.
-----Original Message-----
From: Steve Baker [mailto:sjbaker1@airmail.net]
Sent: Wednesday, August 09, 2000 7:11 PM
To: Sattler Chris-QA1406
Subject: Re: Autonomous Robot
Sattler Chris-QA1406 wrote:
>
> No' I don't think you correctly understood what I am proposing. The
> reflective surface would be on the outside of the funnel shaped piece with
> the narrow tip of the funnel resting on the center of the sensor.
>
> Funnel
> \ / light beam
> \ /______________
> \ /|
> ______V__
> Sensor
>
> The light beam would have to arrive parallel to the sensor. If the outside
> surface of the funnel/cone was at a 45 degree angle to the sensor the • sensor
> would have to be as large around as the widest part of the cone and this
> would determine how much the height of the beam above the sensor could • vary.
> By curving the surface of the cone you could change this do some degree.
>
> Getting the light beam to arrive parallel to the sensor would be a problem
> in and area of uneven terrain. I suppose you could use some kind of system
> to keep the sensor horizontal.
Right - but what I've been trying to say is that once you decide to have a
'tower' with lasers on it - and have it send data to the robot - you might
as well but the light sensors on the tower too - which cuts down on the
weight/power/RCX-outputs needed in the robot - which is "A Good Thing".
Hence the interest in retroreflectors. If you can shine a laser from the
tower and hit the retroreflector, the tower's light sensor (bolted right
next
to the laser and parallel to it) will get a strong return. If the tower
laser
misses the robot - then no reflection.
So the software in the tower has to simply scan the laser around until it
finds the robot - then track it (an easy problem to solve). The tower
can tell the robot where it is in the world VERY easily. You could even
put multiple towers around your home - and use them to track multiple
robots in parallel.
Once you get that right - all your robots know where they are all the time.
If they need to know their heading, they can infer that from their
consecutive
positions relative to the tower when driving in a straight line - and use
odometry when they go around corners or get into a position where their
IR sensor cannot get the position data report from any nearby tower.
This scheme would be *great* because you could install it permenantly
in your home - and just thave all your robots use it. Robotic navigation
then needs ZERO RCX inputs or outputs - and the software on the robot to
read it is very non-intrusive.
--
Steve Baker HomeEmail: <sjbaker1@airmail.net>
WorkEmail: <sjbaker@link.com>
HomePage : http://web2.airmail.net/sjbaker1
Projects : http://plib.sourceforge.net
http://tuxaqfh.sourceforge.net
http://tuxkart.sourceforge.net
http://prettypoly.sourceforge.net
|
|
Message has 1 Reply:
 | | Re: Autonomous Robot
|
| in article BB60654DFAA8D311B164...g.mot.com, Sattler Chris-QA1406 at lego-robotics@crynwr.com wrote on 8/10/00 6:00 AM: (...) Not so good, I would think. The cylindrical surface spreads the beam out and it gets quite a bit weaker and difficult to (...) (24 years ago, 10-Aug-00, to lugnet.robotics)
|
2 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
