Loading ...
Sorry, an error occurred while loading the content.

15916RE: [SeattleRobotics] Re: SRF08 beam angle

Expand Messages
  • Alejandro Alonso Puig
    Dec 9, 2003
      As promised, here you have the results of my testing with the SRF08 sonar
      ranger during the weekend (monday was also holiday at Madrid/Spain :-)

      1) The SRF08 follow more or less the beam pattern at
      http://acroname.com/robotics/parts/R145-SRF08.html
      2) The beam could be forced to be narrower by putting some foam around. I
      have reduced it 20 degrees max. If it is used excessive foam surface or try
      to narrow too much the beam, the results are unpredictable. Sometimes the
      transducer detect the foam and sometimes it does not detect even the
      obstacles just in front. In any case, by this trick it is possible to
      prevent detecting the floor when we do not want it.

      I have finished a technical document on a sonar ranger to understand how
      sonar rangers "see" the surroundings. Unfortunately it is in Spanish, but
      there are several figures that could help understanding. Also there are
      several links in English at the bottom of the page:
      http://mundobot.com/tecnica/Sonar/Sonar.htm

      Whenever I have enough time I'll translate my webpage to English

      Best regards,
      Alejandro Alonso Puig
      http://mundobot.com
      alejandro.alonso@...




      -----Original Message-----
      From: Alejandro Alonso Puig [mailto:alejandro.alonso@...]
      Sent: martes, 02 de diciembre de 2003 17:52
      To: SeattleRobotics@yahoogroups.com
      Subject: RE: [SeattleRobotics] Re: SRF08 beam angle


      Yes, I thought the same as you, eliminating the width in the image in order
      to narrow the figures to its real form. I saw this article. In fact I'm
      working in something similar. Look at my web page site. Its in Spanish yet,
      but I'll treanslate to English soon. The page is not finished, but at least
      you will find a figure that represents what I'm doing with the sonar
      ranger(Hardware and Software): http://mundobot.com/tecnica/Sonar/Sonar.htm

      Best regards

      Alejandro Alonso Puig
      http://mundobot.com
      alejandro.alonso@...




      -----Original Message-----
      From: dan michaels [mailto:dan@...]
      Sent: martes, 02 de diciembre de 2003 17:24
      To: SeattleRobotics@yahoogroups.com
      Subject: [SeattleRobotics] Re: SRF08 beam angle


      --- In SeattleRobotics@yahoogroups.com, "Alejandro Alonso Puig"
      <alejandro.alonso@m...> wrote:

      You might also check out the following page, which shows a sweep
      using such a sonar. The "image" shows extended widths of objects in
      azimuth.

      http://www.hazmat.com/~mjb/projects/picsonar/

      I had often wondered about the idea of doing an inverse spatial
      convolution on such an image in order to try and remove the beam
      width characteristic. This might be a lot of processing for a small
      microcontroller.

      - dan michaels
      ======================


      > Really interesting answers. Thanks. Dan, the graph from Acroname is
      really
      > clear. I'll make some tests during this weekend but it seems that
      the tests
      > I performed more or less agree with the graph. The measures were
      taken at 1
      > meter from the transducer. That is 3,28 feet and as from the
      graph, the
      > width at such distance is of about 80 degrees (not so far from the
      100
      > degrees I measured, considering I'm not testing in an ideal
      environment).
      >
      > In any case I think it is very wide if we really want to have some
      clear
      > information on the surroundings. As per trigonometry
      > (Arc=angleº(pi/180)Distance), the chord (arc) detected at 1 meter
      from the
      > transducer is Arc=80º(pi/180)1= 1,4 meters. That means 0,7m to the
      right,
      > 0,7m to the left, 0,7m to the top and 0,7m to the bottom. If I'm not
      > mistaken, this means that if you want to detect something at a
      distance of 1
      > meter from your robot, the transducer should be either installed at
      least
      > 0,7m from the floor and in 90º position, or install it in a lower
      position
      > but with a vertical inclination greater than 90º, so you will not
      detect the
      > floor (kind of reflecting floor), but you will detect things over
      the robot
      > (may be a problem in some cases)
      >
      > If I am right with this assertions, the SRF08 is good because of
      the I2C
      > interface, but not so good because of the beam width.
      >
      > I read something about putting some kind of foam around the
      transducer to
      > force a reduction on the detection cone. I'm not sure if it will
      work, but
      > I'll try it during the weekend. I'll tell you.
      >
      > By the way, I raised the question to Devantech Company and the
      answer I got
      > was: "Beam width on the SRF04/8 is a function of the transducers
      and is
      > fixed. For the transducers used on the SRF04/8 this is 55 degrees at
      > the -6db point". Not as interesting as your answers.
      >
      > Best regards
      >
      > Alejandro Alonso Puig
      > http://mundobot.com
      > alejandro.alonso@m...
      >
      >
      >
      >
      > -----Original Message-----
      > From: dan michaels [mailto:dan@o...]
      > Sent: martes, 02 de diciembre de 2003 3:48
      > To: SeattleRobotics@yahoogroups.com
      > Subject: [SeattleRobotics] Re: SRF08 beam angle
      >
      >
      > --- In SeattleRobotics@yahoogroups.com, "Phil" <phil1960us@y...>
      > wrote:
      > > what about some sort of tuning "pipe" that you slip over the
      > reciever
      > > to decrease the field of view that the receiver actually sees?
      > sort
      > > of like blinders for a horse...
      >
      >
      > I was thinking about the same sort of thing, but I really haven't
      > much idea how to do it. Sound works differently from light. Sound
      can
      > go around corners, etc. Something to try might be tubes of rolled
      > cardboard/plastic/etc around the transmitter with lengths at "odd"
      > multiples of "quarter" wavelength. At even multiples, I think the
      > sound pressure will be a minimum, but I'm not really sure about this
      > stuff at all. Or possibly a megaphone/conical horn - but again it
      > probably needs the correct length and end aperture size for best
      > results. Good area for experimentation - plus some preliminary
      > legwork into acoustics. Tell us how it goes ;-).
      >
      >
      > - dan michaels
      > www.oricomtech.com
      > ============================
      >
      >
      >
      > Visit the SRS Website at http://www.seattlerobotics.org
      >
      > To unsubscribe from this group, send an email to:
      > SeattleRobotics-unsubscribe@yahoogroups.com
      >
      >
      > Your use of Yahoo! Groups is subject to
      http://docs.yahoo.com/info/terms/


      Visit the SRS Website at http://www.seattlerobotics.org

      To unsubscribe from this group, send an email to:
      SeattleRobotics-unsubscribe@yahoogroups.com


      Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
    • Show all 17 messages in this topic