Jump to content

How to use "Position Averaging"


Gill & Tony

Recommended Posts

I'm planning a holiday in Egypt and have been looking at earthcaches and virtual caches.  This one has me stumped.  The description says, in part

Located on the Giza Plateau. The given coordinates represent one of many possible starting points. This virtual cache will provide you an opportunity to learn about and employ the seldom-used position-averaging feature on your GPS.

The objective of this virtual cache is to circumnavigate Kufu’s pyramid (the tallest structure in the world for over four millennia) then report back with your estimate of the coordinates of its 137 m peak.

To validate your visit, please send me both your track log and the coordinates that you acquired utilizing the position-averaging feature on your GPS. These coordinates should represent your estimate of the actual coordinates of the top of the Kufu’s (Cheops) pyramid.
 
So, I have to walk around the pyramid with my GPS in hand.  This will generate a track log stored somewhere in my GPS and I can then use that to calculate the coordinates of the centre-point of my track using a "Position-averaging" technique.
 
There are so many questions here that I'm not really sure where to start.  Given that I have a Garmin E-Trex 30x.
 
1. How to I start and stop a track log, so I only get the required data in the track log?
2. Where is the track log stored and how do I tell which track log is the one I want - assuming there are other track logs there?
3.  How do I do this "position-averaging" to get the answer?
 
Any help will be greatly appreciated.
 
Thanks
 
Tony
Link to comment

Re 1: Make use of your unit's manual page 14.
Re 2: Tracklogs are stored in [unit drive letter]/Garmin/GPX/Archive
You may create an isolated track by using the stopwatch. Starting and stopping the stopwatch creates an "Stopwatch track" with just the portion you'll need (at least the Montana's stopwatch is behaving this way. Test it on your eTrex 30 (the manual doesn't pay attention on this feature).

8b316c7d18a078bbb5baab0c7a1a1e30.png

Re 3: Make use of your unit's manual page 10.

The manual: http://static.garmin.com/pumac/eTrex_10-20-30_OM_EN.pdf

Hans

Edited by HHL
typos
Link to comment

Hi, Hans

Thanks for that help.

Points 1 and 2 I now understand.

Point 3 doesn't seem to be covered on that page of the manual.  Page 10 covers "Waypoint Averaging" which is a technique for getting an accurate set of coordinates for the current location.  

"Position Averaging" seems to be a technique for getting a location out of a track.  Since I will be walking around a centre-point the track will surround the centre point and this technique seems to be a way to find the coordinates of that centre point, without actually going there.

Since the cache description refers to the "seldom-used" position averaging feature, i suspect that it doesn't mean "Waypoint averaging", since that is quite often used by cache owners.

Thanks again

Tony

Link to comment
1 hour ago, Gill & Tony said:

The objective of this virtual cache is to circumnavigate Kufu’s pyramid (the tallest structure in the world for over four millennia) then report back with your estimate of the coordinates of its 137 m peak.

To validate your visit, please send me both your track log and the coordinates that you acquired utilizing the position-averaging feature on your GPS. These coordinates should represent your estimate of the actual coordinates of the top of the Kufu’s (Cheops) pyramid.

 

The easiest way would be to take coordinates at the 4 corners of the pyramid and then calculate (there are apps/websites to do this) the intersection of the two lines between opposite corners.

I haven't tried this myself but I guess if you mark a WP at the start and use "average" on your GPS and keep that running while walking around the pyramid you would also get the same result.

 

Link to comment

The corner method is a sure bet. You can try the other method of averaging during a walk, but I have a feeling that the GPS won't reach convergence until you are standing still, in which case the averaging will be weighted toward that spot where you spend the most time. It would be a neat exercise to try if you have the time, but I wouldn't rely on it.

Link to comment

You could

  • Mark a carefully averaged waypoint at each corner.
  • Define a route between the corners such that it crosses itself at the peak.
  • Zoom in on the crossing point and read the coords.


I've done this with my old Garmin 60Cx.  Worked well enough to find a cache.


But for small distances, Red90's approach is fine.

Edited by Viajero Perdido
Link to comment

When I read the description, I thought the CO wanted the cacher to go to a corner and average waypoint A.  The go to the next corner and again average waypoint A.  Repeat for the remaining two corners at which time the averaged waypoint A should be close to the coordinates of the center of the pyramid.  There is little value, if any, in averaging each corner individually. 

Link to comment
1 hour ago, GeoTrekker26 said:

When I read the description, I thought the CO wanted the cacher to go to a corner and average waypoint A.  The go to the next corner and again average waypoint A.  Repeat for the remaining two corners at which time the averaged waypoint A should be close to the coordinates of the center of the pyramid.  There is little value, if any, in averaging each corner individually. 

This is in fact the simplest method, and requires no math outside of the GPSr.

Link to comment
1 hour ago, Atlas Cached said:

Hmmmm.... Yes...... Find waypoint "A", use information there to find waypoint "B"... "C".... "D"..... etc, and at each wayoint, use the current location to average the "Final" waypoint.

I was thinking more along the lines of: 
Go to this location and find this object. Take a waypoint reading.
Go to a second location and find this object. Take a waypoint reading.

...

go to the nth location and find this object. Take a waypoint reading.

Average them. The final location is at this average.

Where the locations, rather than given as coordinates on the website, are addresses or generic locations or well known landmarks with objects that should be easy to describe and find as the source to take coordinate readings. As long as the final coordinates lead to a fairly geosensical spot, it'll be challenging but not impossible. It would certainly be more work to set up than to solve.

Edited by Mineral2
Link to comment
1 hour ago, Mineral2 said:

I was thinking more along the lines of: 
Go to this location and find this object. Take a waypoint reading.
Go to a second location and find this object. Take a waypoint reading.

...

go to the nth location and find this object. Take a waypoint reading.

Average them. The final location is at this average.

Where the locations, rather than given as coordinates on the website, are addresses or generic locations or well known landmarks with objects that should be easy to describe and find as the source to take coordinate readings. As long as the final coordinates lead to a fairly geosensical spot, it'll be challenging but not impossible. It would certainly be more work to set up than to solve.

Absolutely! Just don't want to publish coordinates for each waypoint/stage online, as many would just do the math and go straight to the final!

Link to comment
On 9/8/2017 at 7:55 AM, on4bam said:

Since the CO didn't log in at gc since July 2016 and has no finds since 2015 chances are you're OK with any answer you might or might not send :ph34r:

 

And I could just go onto Google Earth and grab some coordinates from there. :rolleyes:

I know for certain that some of my virtual/earthcache finds have had inactive CO's, but I do like to at least try and do what was asked for.

  • Upvote 1
Link to comment

There's a learning opportunity here (beyond the easy hook) so thanx for at least thinking it through.

There's an "obvious" problem with the "take the median of the four corners to find the 'center' of the perimeter" approach. What is the problem?
If you don't know, think it through. Share your thoughts knowing that there's a moderator that's asked you and that'll have your back for people being mean.
If you know - helpful, coaching answers only please (and only in encouraging response to others that are standing on their tip toes to stretch and grow)
If you're a flat earther, please don't participate. 

It's also a good reminder of several things wrong with virtual caches. Those are explicitly outside the charter of this group, so please refrain from that discussion.

Link to comment

The CO specifically mentions "the seldom-used position-averaging feature on your GPS."  I am taking this to mean something other than the waypoint averaging feature, since most CO's would be well aware of this and also taking it to mean something other than just grabbing 4 waypoints and calculating the diagonal intersection.

I had assumed that the track log could somehow be used to calculate the centre point of the area.  I know it can be used to map the area enclose (in SqM) but I can't see how to get the centre point.

I have plenty of time to learn (I'm not leaving until August next year) but my plan, if all else fails, is to grab coordinates from Google Earth, take waypoints at the 4 corners and calculate the intersection using the GE coordinates as a sanity check.

I would, however, like to do it the way the CO intended, by learning about this feature assuming the E-Trex 30 has the feature installed.

Link to comment
9 hours ago, robertlipe said:

There's a learning opportunity here (beyond the easy hook) so thanx for at least thinking it through.

There's an "obvious" problem with the "take the median of the four corners to find the 'center' of the perimeter" approach. What is the problem?
If you don't know, think it through. Share your thoughts knowing that there's a moderator that's asked you and that'll have your back for people being mean.
If you know - helpful, coaching answers only please (and only in encouraging response to others that are standing on their tip toes to stretch and grow)
If you're a flat earther, please don't participate. 

It's also a good reminder of several things wrong with virtual caches. Those are explicitly outside the charter of this group, so please refrain from that discussion.

Immediately coming to mind is the potential for the corners to not all be equidistant from the center, and/or the object may not be perfectly square, and/or the center of the object may not actually be in the exact center of the perimeter. 

Link to comment
11 hours ago, robertlipe said:

There's a learning opportunity here (beyond the easy hook) so thanx for at least thinking it through.

There's an "obvious" problem with the "take the median of the four corners to find the 'center' of the perimeter" approach. What is the problem?
If you don't know, think it through. Share your thoughts knowing that there's a moderator that's asked you and that'll have your back for people being mean.
If you know - helpful, coaching answers only please (and only in encouraging response to others that are standing on their tip toes to stretch and grow)
If you're a flat earther, please don't participate. 

It's also a good reminder of several things wrong with virtual caches. Those are explicitly outside the charter of this group, so please refrain from that discussion.

I'm not really seeing the obvious problem.
First, you want to take the mean, not the median. It probably won't make a difference here, but a mean is more correct for this application.

Second, we're dealing with the pyramids at Giza. I might agree that in general, the mean of a cluster of points isn't necessarily the center of the perimeter. But in the case of perfect shapes, the average of the vertices would represent the center. And we know that the base of the pyramids is approximately a perfect square. Therefore the mean of the vertices is an adequate estimate of the summit of the pyramid. 

Third, if we're willing to throw out the assumption of a perfect square (or a rectangular base), then we should also throw out the assumption that the summit of the pyramid is located along the lines that connect the opposite corners, in which case we'll need more architectural information about the structure to make our calculation. That's beyond what is asked of this cache.

Fourth, There is also a problem in using the perimeter, whether keeping waypoint averaging going along the walk or recording a track and averaging at home: That problem is that if you don't walk at a constant speed, your average will be weighted toward points where your speed was slower. The only exception to this is collecting track information by distance rather than time or auto. Even this has its limits, especially if there are obstacles that prevent you from travelling in a straight line along the base of the pyramid from corner to corner.

At this scale, I'm not concerned with the curvature of the earth causing problems. And if you're concerned that the scale of the latitude and longitudinal axes aren't equal, then we can work in the UTM coordinate system, which I recommend anyway.
 

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...