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

8366Re: Red Dot vs Dead Reckoning + More

Expand Messages
  • apenadragon
    Apr 1, 2011
    • 0 Attachment
      ok, that invovled more math than my sleep deprived mind can handle at the moment, will try it again tomorrow when I've had some sleep.

      regardless, I've got my red dot turned off as it doesn't give me anything useful for my purposes.

      If I can get the new Dev version before I leave at 3pm I'll have the YL see what it does as we drive home.


      and James, thanks for your patience in trying to educate me. I'm playing catch up on some of this stuff.

      --- In aprsisce@yahoogroups.com, "Lynn W Deffenbaugh (Mr)" <kj4erj@...> wrote:
      > Very well explained James. I have isolated the problem, implemented the
      > solution, and am about to release a new Dev version for those of you
      > that can (safely) see your screen while mobile can test (and hopefully
      > verify) my correction. I base the Dead Reckoning of ME on whether or
      > not you have Configure / Beacon / CourseSpeed checked. If so, ME will
      > dead reckon based on the (hopefully) same information a recipient of
      > your last beaconed packet would be using. If you don't beacon
      > CourseSpeed, then I don't DR ME since you KNOW where you are and where
      > you're heading and you haven't told anyone else anyway.
      > As for the difference/similarity of ME's DR vs the Red Dot.... If you
      > measure the deviation between ME's actual position and the corresponding
      > DR symbol, scale that according to the Configure / Genius / Forecast
      > Error setting, you'll have a bearing (from actual to DR) and a
      > percentage of error (distance / Forecast Error * 100%). Plot that
      > value on the circle based on the compass points where touching the
      > circle = 100% and you have the red dot, a Visualization of the range and
      > direction of the error between a time-based extrapolation of your last
      > beaconed position, speed, and course vs your actual current position, in
      > other words, the red dot displays the magnitude and direction of the
      > error between where an outside observer thinks you are based on the time
      > since and values within your last beacon and where you know yourself to
      > actually be.
      > If the DR is on top of you (constant, straight line motion), the red dot
      > will be on top of you and invisible. You won't see the DR symbol
      > either, most likely.
      > If you speed up (still in a straight line), the DR symbol will start
      > falling behind, and the red dot will start drifting towards the circle,
      > lagging directly behind your current direction of motion, but the DR and
      > red dot will be at different rates because the red dot is a percentage
      > of the allowed Forecast Error (when it touches the circle), and the DR
      > will be where that outside observer would think you are. So, the DR
      > moves according to map scale and the red dot moves according to the
      > magnitude of the allowed Forecast Error.
      > If you slow down (still in a straight line, we can do that here in
      > Florida for long distances), the DR symbol will start pulling out in
      > front of your direction of motion, and so will the red dot, again at
      > different rates. If you take the distance between your actual position
      > and the DR symbol, divide it by the configured Forecast Error, that's
      > relatively how far the red dot will be between the center and edge of
      > the circle, still in your direction of motion.
      > I'll leave it as an exercise for the reader to figure out where the DR
      > and red dots will go if you vary just the direction and maintain speed
      > (should drift away from you at 90 degrees) or if you vary both speed and
      > direction.
      > For extra credit, you can play the "red dot game" (yes, this is
      > absolutely worse than just glancing at the APRSISCE/32 screen while
      > driving). Purposely vary your speed and/or direction to control where
      > the red dot is on the circle. Hint: it's a lot easier if you have a
      > very long stretch of very straight road and works fairly well at giving
      > you something to do to stay alert, even if not truly looking where
      > you're going). Hint #2: it's highly discouraged to attempt to control
      > the red dot in any orientation other than directly ahead or behind your
      > current heading unless you're driving on a salt flat or something.
      > Think about it....
      > Now, to quit typing, do a few release notes entries, and get this
      > development release out there for testing!
      > Lynn (D) - KJ4ERJ - Author of APRSISCE for Windows Mobile and Win32
      > On 4/1/2011 1:52 PM, James Ewen wrote:
      > > On Fri, Apr 1, 2011 at 10:35 AM, apenadragon<kc2ant@...> wrote:
      > >
      > >>> So, I just need to fix ME's DR to use the last beaconed speed/heading
      > >>> rather than the current GPS speed/heading and all will be well.
      > >> Do you?? personally I don't see the issue... the DR is a projection of where
      > >> the station "should" be...
      > > Actually Adam, it sounds like you're missing the issue.
      > >
      > > If you take the last reported speed and heading and project that from
      > > the last reported location, as time progresses, the dead reckoned icon
      > > will move to where the station "could" be.
      > >
      > > The problem with what we are seeing is that Lynn is using the CURRENT
      > > speed and heading, and projecting a possible location using that
      > > information from the last reported location.
      > >
      > > Lets suggest that a position was reported, with a speed of 60 mph, and
      > > a heading of 90 degrees. After one minute, the dead
      > > reckoned position should be 1 mile east of the reported location.
      > >
      > > The way it is implemented, if I stop driving east after 30 seconds,
      > > the DR position would return back to the last reported location. If I
      > > were to turn south and drive for 1 minute at 60 mph (without sending
      > > an updated position report), there would be a DR object 1 mile south
      > > of the originally reported position.
      > >
      > > Now, if this instantaneous line were to originate from the current
      > > location, it would still be wrong. The line leaving my current
      > > location would not be a dead reckoned position, but rather a projected
      > > location for a future position report based on the current speed and
      > > course, projected into the future the amount of time since the last
      > > reported position.
      > >
      > > When Lynn gets it straightened out, you'll see something interesting
      > > happening. As Lynn said, if you continue in a straight line at a
      > > constant speed, you won't see the projected icon. If the road turns,
      > > or you change speed, then you'll see the projected (DR) icon continue
      > > along the last reported vector. Once the Genius settings determine
      > > that you've strayed too far, a new position report will be sent, and
      > > the DR icon will start moving away from the reported position again.
      > >
      > > James
      > > VE6SRV
    • Show all 25 messages in this topic