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

Re: [RoboMower] Re: Friendly Machines & Stiga / Toro

Expand Messages
  • wolfbob
    There are a number of personal GPS-like systems for robots working inside warehouses and such. They can locate to less than a cm. Surveyors use special
    Message 1 of 34 , Nov 4, 2009
    View Source
    • 0 Attachment
      There are a number of "personal" GPS-like systems for robots working inside warehouses and such. They can locate to less than a cm. Surveyors use special differential GPS units and have accuracies of a tenth of an inch of better at a 1/4 mile. Do a google search for precision GPS. Commercial aircraft have sub cm 3D capability for landing at major airports. Developed by Interstate Electronics some 15 years ago. The technology exists, it is just not too applicable for you back yard (yet).

      WBob
      ----- Original Message -----
      From: jabraham@...
      To: RoboMower@yahoogroups.com
      Sent: Tuesday, November 03, 2009 3:49 PM
      Subject: [RoboMower] Re: Friendly Machines & Stiga / Toro


      Interesting. Thanks for the info. Of course it's technically feasible to triangulate off of three local transmitters (not sure what the other poster meant by "obstacles" -- lots of RF waves go through obstacles.)

      But since you've pointed out the cost is in the accurate timing, and you'd need accurate timing regardless of whether you're tracking points in the sky or points around your house, it would make more sense to mount a differential transmitter at a fixed point on your house (to correct for atmospheric interference) and use points in the sky instead of the three points on your property your house.

      Regardless, it sounds very expensive.

      What about the ultrasound idea, to avoid the expensive timer? Is there no off-the-shelf technology for measuring distance from a fixed transmitter using ultrasound? That sounds more promising than RF. Might drive the pets crazy though.

      --
      John

      --- In RoboMower@yahoogroups.com, Danny Miller <dannym@...> wrote:
      >
      > Actually, generally no. Because the speed of light is extremely fast,
      > most of the problems are in timing the signal and have little to do with
      > the satellite being so far away. There's no good way to make consumer
      > transmitters with the accuracy required, in general it requires an
      > atomic clock for one because the satellite transmission time must be
      > sync'ed to picoseconds. And running a clocking cable from one
      > transmitter to the other to guarantee they all have the same clock, even
      > that is fraught with problems. Never mind that a GPS-like but
      > not-actually-GPS receiver would be a highly specialized device which
      > would be incredibly expensive to design.
      >
      > Ultrasonic transceiver have some potential because the speed of sound is
      > so much slower that distance results in far more discernible delays, but
      > cycle-accurate reception at a distance is very difficult.
      >
      > Actually there ARE survey gps units which are accurate to within
      > millimeters. It's POSSIBLE. There's a major cost issue, however. The
      > units have several features such as ultra-accurate clocks which are not
      > present in consumer units because the cost is orders of magnitude higher.
      >
      > Danny
      >
      > jabraham@... wrote:
      > > --- In RoboMower@yahoogroups.com, "WB" <wild_bill_howell@> wrote:
      > >
      > >> I don't believe that GPS has the fine resolution required for mowing where an error of just a few inches could place the mower in the path of road traffic in yards w/o sidewalks.
      > >>
      > >>
      > >
      > > No it doesn't. Another alternative is to bolt three fixed transmitters around the house and triangulate off of them. That would give enough accuracy. But I wonder whether 1) it would be too expensive to develop the hardware/software, and 2) whether bolting three fixed transmitters to your house and powering them is any easier than running a perimeter wire?
      > >
      > > --
      > > John Abraham
      > >
      > >
      > >
      >





      [Non-text portions of this message have been removed]
    • rowell@mindspring.com
      Yes slip would make odometer style dead reconing unlikely in something in such a rough environment. RFID tags in conjunction with dead reconing could
      Message 34 of 34 , Nov 7, 2009
      View Source
      • 0 Attachment
        Yes slip would make odometer style "dead reconing" unlikely in something in such a rough environment. RFID tags in conjunction with dead reconing could possibly compensate for slippage though by confirming the bot's location upon crossing a tag.

        These would be easier than laying out a wire because they don't have to be contiguous, dont require power, etc. One would just place them stragecially about the lawn, perhaps closer togeather in curved areas, run some kind of calibration routine where you manually guide the bot around the perimeter, then save the configuration.

        keith rowell

        --- In RoboMower@yahoogroups.com, "Dan Barclay" <Dan@...> wrote:
        >
        > Mapping off tire size/odometer won't work at all in my yard. Mud here and
        > there, enough for it to slip. It needs actual position or perimeter.
        >
        >
        >
        > As for "invisible walls", they need transmitters. That would be far more
        > trouble than the perimeter wire in my lawn and most others I've looked it.
        > Lots of curved edges and no place to put a transmitter along the pool edge.
        >
        >
        >
        > Dan
        >
        >
        >
        > From: RoboMower@yahoogroups.com [mailto:RoboMower@yahoogroups.com] On Behalf
        > Of Ben Sporl
        > Sent: Tuesday, November 03, 2009 9:10 AM
        > To: RoboMower@yahoogroups.com
        > Subject: Re: [RoboMower] Re: Friendly Machines & Stiga / Toro
        >
        >
        >
        >
        >
        > I agree, pure GPS is too general. Hopefully they'll upgrade the satellite
        > system in the near future to correct for this variance.
        >
        > However, you don't need GPS if you have a good mapping program. Using fixed
        > tire sizes, you can map how far the mower has moved and the direction. Using
        > better algorithms they can definitely comp up with an option.
        >
        > Maybe take a note from iRobot's playbook and deploy solar powered invisible
        > walls. They charge during the day, then turn on when the mower runs. Make
        > them laser to control the line allowing highly directional wall.
        >
        > Maybe put some floaties to keep the mower from accidently sinking to the
        > bottom of the pool, and an emergency kill switch in case it pulls out the
        > hidden lasers and starts attacking.
        >
        > ________________________________
        > From: WB <wild_bill_howell@... <mailto:wild_bill_howell%40yahoo.com> >
        > To: RoboMower@yahoogroups.com <mailto:RoboMower%40yahoogroups.com>
        > Sent: Mon, November 2, 2009 4:44:04 PM
        > Subject: [RoboMower] Re: Friendly Machines & Stiga / Toro
        >
        > I don't believe that GPS has the fine resolution required for mowing where
        > an error of just a few inches could place the mower in the path of road
        > traffic in yards w/o sidewalks.
        >
        > WB
        >
        > --- In RoboMower@yahoogrou ps.com, Ben Sporl <bsporl@> wrote:
        > >
        > >
        > >
        > > Thanks for the article.
        > >
        > >
        > > It's interesting to see a bit insight into what they were thinking. Yes,
        > 10% of the 7m mower market would be a good goal, but not when your mower
        > costs 8x the volume push mowers. I truly think three things need to occur
        > before robomowing takes off:
        > > - Easier: They need to get rid of the perimeter wire and use a mapping
        > application. GPS is built into every phone, car, etc... and can be had for a
        > few bucks when volume ordering. Software can map the yard to ensure the
        > mower is kept in a zone. Even if GPS is not feasible, algorithms can map the
        > yard if you drive the mower around the perimeter. They need to call Google,
        > offer them $1m, and ask for a mapping application in return.
        > > - Lower the costs: They need to be realistic about the costs and not
        > expect people to pay $1,500 when they can buy a quality push mower for $200
        > > - Local support: Local dealerships or partnerships need to be established.
        > For this to work, the customer need to be able to return problem mowers to
        > Sears, Lowes, Ace, Home Depot, etc... Customers need that warm and fuzzy
        > feeling of a local person.
        > >
        > > Anyway, that's my 2 cents.
        > >
        > >
        > >
        > >
        > > ____________ _________ _________ __
        > > From: WB <wild_bill_howell@ ...>
        > > To: RoboMower@yahoogrou ps.com
        > > Sent: Sat, October 31, 2009 12:06:15 PM
        > > Subject: [RoboMower] Re: Friendly Machines & Stiga / Toro
        > >
        > >
        > > This is somewhat old news (2006 Design News), but answers a few questions
        > and reveals some interesting history of the Friendly RoboMower. :
        > >
        > > http://www.designne ws.com/article/ 9955-Mowing_ on_Autopilot. php
        > >
        > > WB
        > >
        > >
        > >
        > >
        > > [Non-text portions of this message have been removed]
        > >
        >
        > [Non-text portions of this message have been removed]
        >
        >
        >
        >
        >
        > [Non-text portions of this message have been removed]
        >
      Your message has been successfully submitted and would be delivered to recipients shortly.