🚀 go-pugleaf

In article <15792@cup.portal.com> MJB@cup.portal.com (Martin J Brown-Jr) writes:
>I need a system that measures distances for some homebrew robotics projects.
>I would like to be able to measure to an 1/8th of an inch
>accuracy or greater, over a distance of 1/4" (or so) to 30 feet. Also, and
>more importantly, to have it be a narrow beam (1/4"?) width spread at the
>maximum distance. This is so that I will be able to get ranges at specific
>spots, and to be able to "see" smaller than wall size objects.

  One way to achieve a smaller effective beam width is through a process known
as "beam splitting".  In order to effectively narrow the beam spread, multiple
sonar transducers are used, and their beams are overlapped.  Distance 
measurements are taken sequentially from each transducer and compared to one
another.  If the distances are all within a certain tolerance of each other,
an object is presumed to be in the area where the beams overlap.  I  believe
this is how the process works.  I have read a number of papers on mobile
robotics where beam splitting is used and the above is what I have gleaned from
them.  A good source of info is the _IEEE Journal of Robotics and Automation_
(pay special attention to the research being done at CMU).
  Another way to get the accuracy and beam width you want is to use narrow beam
sonar transducers available from various manufacturers.  (Can't recall where I
saw these right now).  Also, don't forget to correct for air temperature when
making sonar measurements.  I don't have the correction formula with me, but
let me know if there is interest, and I will post it or mail it.
  I'm working on a mobile robot project of my own at home also.  Currently I
am designing the sensor head of the robot and I am working on a beam splitting
arrangement using Polaroid transducers.  I would like to see more discussion
on robotics here, also feel free to E-mail me.

							Steve

Steve Kohls
skohls@uceng.UUCP