To LUGNET HomepageTo LUGNET News HomepageTo LUGNET Guide Homepage
 Help on Searching
 
Post new message to lugnet.roboticsOpen lugnet.robotics in your NNTP NewsreaderTo LUGNET News Traffic PageSign In (Members)
 Robotics / 2629
2628  |  2630
Subject: 
Re: cool idea for all-around range sensor
Newsgroups: 
lugnet.robotics
Date: 
Mon, 11 Jan 1999 00:04:35 GMT
Original-From: 
Mark Williamson <markw98@ibm.&Spamcake&net>
Viewed: 
3645 times
  
sounds interesting.  i figured i was oversimplifying it.  but i've seen bots
with 24 sonar sensors around the machine.  since most controllers dont have
24 input ports, there's probably a round robin action going on, pulsing
between each of the 24 units, with the timing sufficiently set so you don't
miss a return on the signal.  there probably are some more chips involved.
but this could be interesting if it works because it's a lot cheaper than
sonar, but perhaps not as reliable.

I just have to dust off my digital circuits book and figure out which chips
to use (decade counters, multivibrators, 556's, etc.  just gotta figure that
part out.)


-----Original Message-----
From: Jim Moran <lugnet.robotics@lugnet.com>
To: lego-robotics@crynwr.com <lego-robotics@crynwr.com>
Date: Sunday, January 10, 1999 3:28 PM
Subject: Re: cool idea for all-around range sensor


> Have you devised some way for the output of each IR receiver to be bistable
> as opposed to analog ?
> 
> In order to wire-or a bunch of them with resistors and get a unique,
> repeatable reading for each receiver being "active" would require two • stable
> states for each element (and nothing in between).
> 
> This is easily done with switches, but I don't think such a scheme will • work
> for IR phototransistors without some more electronics between them.
> Varying IR light levels will not always saturate the receiver elements
> (assuming that was what you were planning to use as your active state), and
> the resistance values you designed in will be compromised.
> 
> A time-shifted scheme might be interesting. Maybe you could setup the
> emitters to fire in a predetermined order at precise intervals. Create an
> "index" element (resistors only) that returns the same fixed level (a level
> not achievable by any of the actual IR receivers) when the firing sequence
> first begins. Syncronize the reading of the rest of the elements based on
> seeing this "sync" level. This would permit you to read a range of values
> from the receivers directly, and you could wire-or them without any
> resistors.
> 
> --
> SeeYa !
> ------------
> Jim
> -----------------------------------------------------------------------
> From the keyboard of Cyberia - (the machine's name, not mine)
> 
> Mark Williamson <lego-robotics@crynwr.com> wrote in message
> news:007d01be3cd4$f4be5440$6119fea9@mark.pacbell.net...
> > i've read all the messages on all-around bump sensing, and have an idea.
> > 
> > I'm not actually a mindstorms owner, but have the MIT Handyboard and • bought
> > some Radioshack IR emitter/detector LED's which i'm making into sensors.
> > I'm using the 1x2 legos to house the sensors (pic attached).  I've come up
> > with an idea on arranging these sensors (5 of them) in a triangular patter
> > to give 180 degree ranging view.  How this works is all of the sensors are
> > tied together in such a way that the values of each one are in steps. • Using
> > different values of resistors on the detectors will give you a range.
> > Starting at the left side, the first sensor gets the lowest (or none) • value
> > resistor.  Moving up to the next sensor, we increment the resistor's • value,
> > etc.  Now by actually wiring all 5 of the IR detectors together, we can • read
> > the entire array with one input on the RCX or Handyboard.  An input on
> > sensor 1 will give a higher value than input on sensor 2, with sensor 5
> > giving off the highest value.  If more than one sensor gets input, you • will
> > still know what "side" the input occurred on by noting that the value was
> > excessive.  Then, just mount the IR emitters on the top of the detectors
> > (using another 1x2 lego), you have an IR 180 degree eyeball.
> > 
> > I'm sure there are flaws in this design; but it sounds good in theory.
> > 
> > Regards,
> > Mark Williamson
> > 
> > 
> > 
> > --
> > Did you check the web site first?: http://www.crynwr.com/lego-robotics
> 
> --
> Did you check the web site first?: http://www.crynwr.com/lego-robotics


--
Did you check the web site first?: http://www.crynwr.com/lego-robotics



1 Message in This Thread:

Entire Thread on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact
    

©2005 LUGNET. All rights reserved. - hosted by steinbruch.info GbR