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

Re: i found some info on a LASER RANGE FINDER DESIGN

Expand Messages
  • dpa_io
    Or use structured light to get a grid rather than a line: like the planetary rovers. -dpa
    Message 1 of 9 , Nov 6, 2011
    • 0 Attachment
      Or use structured light to get a grid rather than a line:

      <http://www.seattlerobotics.org/encoder/200110/vision.htm>

      like the planetary rovers.

      -dpa




      --- In SeattleRobotics@yahoogroups.com, "Peter Balch" <peterbalch@...> wrote:
      >
      > From: Weston Turner
      > > You can do the same thing with a webcam, duct-tape, and a laser pointer.
      >
      > Better still, use a laser that has a diffraction grating in front of it to
      > produce a line
      > http://www.ebay.co.uk/itm/Handheld-Laser-Red-Pointer-Water-Surface-Level-Leveler-Leveller-Indicator-Meter-/330637293667?pt=LH_DefaultDomain_15&hash=item4cfb84b863#ht_4932wt_1139
      >
      > http://www.ebay.co.uk/itm/New-Laser-Level-Projection-Aligner-Horizontal-Line-/180673884302?pt=LH_DefaultDomain_0&hash=item2a1100b08e#ht_3143wt_1185
      >
      > So you can get a whole strip of distances - not just a single point.
      >
      > Peter
      >
    • Peter Balch
      From: dpa_io ... To me, that looks like a line rather than a grid of dots ... AFAIK, Pathfinder and Spirit/Opportunity all use stripes. I think they have 5
      Message 2 of 9 , Nov 6, 2011
      • 0 Attachment
        From: dpa_io
        > Or use structured light to get a grid rather than a line:
        > http://www.seattlerobotics.org/encoder/200110/vision.htm

        To me, that looks like a line rather than a grid of dots

        > like the planetary rovers.

        AFAIK, Pathfinder and Spirit/Opportunity all use stripes. I think they have
        5 laser stripes that can be turned on/off. The camera takes a picture with
        none on then another picture with one of the lasers on and compares them.

        I remember seeing a multiple-stripe system back in the 90s. The problem is
        that if you have lots of stripes then either a displaced stripe can appear
        in the "domain" of the next stripe along. Or the laser and camera are close
        so that can't happen and you get much lower resolution.

        I suggested that they use a grid (i.e. x-stripes and y-stripes) with the
        angular-displacemnt in one axis one tenth of the angular-displacemnt in the
        other axis. That way, the big displacement provides the resolution but the
        small displacement provides disambiguation. Don't know if they ever did it.

        Peter
      • jamericanfreddy
        from this designs i see no IR PASS FILTER SET AT 950MM OR 980MM WAVE LENGTH OF most lasers,outside on direct sunlight or inside lighting can be a problem ,when
        Message 3 of 9 , Nov 7, 2011
        • 0 Attachment
          from this designs i see no IR PASS FILTER SET AT 950MM OR 980MM
          WAVE LENGTH OF most lasers,outside on direct sunlight or inside lighting can be a problem ,when wecam doesnt have any filters

          --- In SeattleRobotics@yahoogroups.com, "Peter Balch" <peterbalch@...> wrote:
          >
          > From: dpa_io
          > > Or use structured light to get a grid rather than a line:
          > > http://www.seattlerobotics.org/encoder/200110/vision.htm
          >
          > To me, that looks like a line rather than a grid of dots
          >
          > > like the planetary rovers.
          >
          > AFAIK, Pathfinder and Spirit/Opportunity all use stripes. I think they have
          > 5 laser stripes that can be turned on/off. The camera takes a picture with
          > none on then another picture with one of the lasers on and compares them.
          >
          > I remember seeing a multiple-stripe system back in the 90s. The problem is
          > that if you have lots of stripes then either a displaced stripe can appear
          > in the "domain" of the next stripe along. Or the laser and camera are close
          > so that can't happen and you get much lower resolution.
          >
          > I suggested that they use a grid (i.e. x-stripes and y-stripes) with the
          > angular-displacemnt in one axis one tenth of the angular-displacemnt in the
          > other axis. That way, the big displacement provides the resolution but the
          > small displacement provides disambiguation. Don't know if they ever did it.
          >
          > Peter
          >
        • dpa_io
          Isn t that what the optical filter described in Diagram does? Depicted symbolically here, the band-pass filter passes the 671nm wave length of interest while
          Message 4 of 9 , Nov 7, 2011
          • 0 Attachment
            Isn't that what the optical filter described in Diagram does?

            "Depicted symbolically here, the band-pass filter passes the 671nm wave length of interest while blocking other spectral content."

            -D


            --- In SeattleRobotics@yahoogroups.com, "jamericanfreddy" <jamericanfreddy@...> wrote:
            >
            > from this designs i see no IR PASS FILTER SET AT 950MM OR 980MM
            > WAVE LENGTH OF most lasers,outside on direct sunlight or inside lighting can be a problem ,when wecam doesnt have any filters
            >
            > --- In SeattleRobotics@yahoogroups.com, "Peter Balch" <peterbalch@> wrote:
            > >
            > > From: dpa_io
            > > > Or use structured light to get a grid rather than a line:
            > > > http://www.seattlerobotics.org/encoder/200110/vision.htm
            > >
            > > To me, that looks like a line rather than a grid of dots
            > >
            > > > like the planetary rovers.
            > >
            > > AFAIK, Pathfinder and Spirit/Opportunity all use stripes. I think they have
            > > 5 laser stripes that can be turned on/off. The camera takes a picture with
            > > none on then another picture with one of the lasers on and compares them.
            > >
            > > I remember seeing a multiple-stripe system back in the 90s. The problem is
            > > that if you have lots of stripes then either a displaced stripe can appear
            > > in the "domain" of the next stripe along. Or the laser and camera are close
            > > so that can't happen and you get much lower resolution.
            > >
            > > I suggested that they use a grid (i.e. x-stripes and y-stripes) with the
            > > angular-displacemnt in one axis one tenth of the angular-displacemnt in the
            > > other axis. That way, the big displacement provides the resolution but the
            > > small displacement provides disambiguation. Don't know if they ever did it.
            > >
            > > Peter
            > >
            >
          • jamericanfreddy
            yes in this article yes http://www.seattlerobotics.org/encoder/200110/vision.htm#sec2 but in others just webcam and laser one when i was checking out bandpass
            Message 5 of 9 , Nov 7, 2011
            • 0 Attachment
              yes in this article yes
              http://www.seattlerobotics.org/encoder/200110/vision.htm#sec2
              but in others just webcam and laser
              one when i was checking out bandpass IR filter at edmund optics prices are very high,plus webcam and then lasers,if you read the article on parallax design the guy who made tested a few lasers, and other tests and uses a special camera and adddd a optic IR bandpass lens
              so what it comes down to is a compete LASER RANGING circuit without getting a laser,plus webcam ,plus OPTICS and build it your self and hope it has good accuracy

              dont think it is good as $300 to $3000 LIDAR system ,but looks better then homemade design for less $$$
              i built the first design link on just a webcam and laser and results was bad,lighting was a problem and webcam video not so great(tracking)
              in second by seattle groups encoder design does look alot better,but at high costs

              --- In SeattleRobotics@yahoogroups.com, "dpa_io" <davida@...> wrote:
              >
              >
              >
              > Isn't that what the optical filter described in Diagram does?
              >
              > "Depicted symbolically here, the band-pass filter passes the 671nm wave length of interest while blocking other spectral content."
              >
              > -D
              >
              >
              > --- In SeattleRobotics@yahoogroups.com, "jamericanfreddy" <jamericanfreddy@> wrote:
              > >
              > > from this designs i see no IR PASS FILTER SET AT 950MM OR 980MM
              > > WAVE LENGTH OF most lasers,outside on direct sunlight or inside lighting can be a problem ,when wecam doesnt have any filters
              > >
              > > --- In SeattleRobotics@yahoogroups.com, "Peter Balch" <peterbalch@> wrote:
              > > >
              > > > From: dpa_io
              > > > > Or use structured light to get a grid rather than a line:
              > > > > http://www.seattlerobotics.org/encoder/200110/vision.htm
              > > >
              > > > To me, that looks like a line rather than a grid of dots
              > > >
              > > > > like the planetary rovers.
              > > >
              > > > AFAIK, Pathfinder and Spirit/Opportunity all use stripes. I think they have
              > > > 5 laser stripes that can be turned on/off. The camera takes a picture with
              > > > none on then another picture with one of the lasers on and compares them.
              > > >
              > > > I remember seeing a multiple-stripe system back in the 90s. The problem is
              > > > that if you have lots of stripes then either a displaced stripe can appear
              > > > in the "domain" of the next stripe along. Or the laser and camera are close
              > > > so that can't happen and you get much lower resolution.
              > > >
              > > > I suggested that they use a grid (i.e. x-stripes and y-stripes) with the
              > > > angular-displacemnt in one axis one tenth of the angular-displacemnt in the
              > > > other axis. That way, the big displacement provides the resolution but the
              > > > small displacement provides disambiguation. Don't know if they ever did it.
              > > >
              > > > Peter
              > > >
              > >
              >
            Your message has been successfully submitted and would be delivered to recipients shortly.