[Therion] Problem with elevation view

Torsten Schnitter torsten.schnitter at netcologne.de
Sat Nov 3 11:30:14 CET 2018


Hi

Concerning the loop closure problem I observed some more points.

Following is the set:
A underwater cave with a GPS position at the start and at the water surface (0m).
Inbetween there is a second surface point with also a GPS position at the water surface (0m).
Then there is a second start of survey from inside the cave. We have a GPS position of one point inside the cave (done with radio location).
Both surveys should come together at one point of course.
Due to underwater survey the x and y coordinates (compass) is not that accurate and the two final points do not match exactly (of course not).

In picture "Cave-Lake-without-loop-closed-plan-view" you can see the plan view from this two points which are not at the same place (as expected).
In picture "Cave-Lake-without-loop-closed-elevation-view" you can see the elevation view from this two points.
May be it's hard to see but both points are at the exact same level (which they should be).

(I draw a small scrap - red line - to show the water level (at correct level 0m) before the loop closure.)

After the loop closure I get some problems.
On the picture "Cave-Lake-with-loop-closed-plan-view" you can see the loop closed and corrected in x and y orientation.
This seems to be ok (no problem with that).
On the picture "Cave-Lake-with-loop-closed-elevation-view" you can see the elevation view after loop closure (it doesn't make any difference using therion or survex as loop closure algorithm).
You can see the water level is not at the 0m level (scrap line) anymore, it's far below that. The loop closure corrected the survey also in z orientation although the two "end" points have been at the same level before and no correction should have been necessary.

Due to this corrections the "lake" inside the cave is not flat anymore but it is bend somehow (high at beginning, lower in the middle and high up again at the end). I assume everybody does agree it's not possible to have a lake with bended surface. ;-)

Due to this corrections I observed another point as well.
Having a vertical shot before the loop closures (and for sure this is vertical!) this shot is not vertical anymore after the loop closures.

In my opinion there are two points:
- Having corrections in x and y with loop closures doesn't mean implicitly there is also a correlating correction to the z axis. Especially if the points are at the same level already.
- With loop closure there should be a special treatment with vertical shots (in x/y orientation) and with horizontal shots (in z orientation). May be there should be no correction due to loop closures!?

What do you think?

Btw:
does anybody have a solution for my problem: No correction in z orientation for this loop closure?

regards,
Torsten

> Torsten Schnitter via Therion <therion at speleo.sk> hat am 1. November 2018 um 21:23 geschrieben:
> 
> 
>     Hi Martin
> 
>     I tried both loop closure methods. Survex and Therion.
>     It doesn't make any difference.
> 
>     With both it's not at one level without the extra 0,01m for vertical shots.
>     Both are correct with the extra for the shots.
> 
>     hth,
>     Torsten
> 
>         > > Martin Sluka via Therion <therion at speleo.sk> hat am 1. November 2018 um 16:27 geschrieben:
> > 
> >         Thorsten
> > 
> >         have you try both loop closure methods - Therion’s and Survex’s?
> > 
> >         Martin S.
> > 
> > 
> > 
> >             > > >             1. 11. 2018 v 16:09, Torsten Schnitter via Therion < therion at speleo.sk mailto:therion at speleo.sk >:
> > > 
> > > 
> > >             Hi
> > > 
> > >             I figuered out the problem and may be there is a bug in loop closure... or somewhere else.
> > > 
> > >             The cave does start at 0m of depth and with a fixed point (fix 0 x y z) with z=118.4m.
> > >             After roughly 800m of underwater cave survey (which does result in absolute z values for every station) the cave does surface again at a depth of 0m.
> > >             I fixed this point as well (fix 0 x y z) with z=118.4m (to correct the x and y error).
> > > 
> > >             So far so good.
> > >             But further (around 1km more) into the cave it does surface again and I realized in elevation view that 0m is not at the same level than the first two points are.
> > >             As soon as I deleted one of the two fixed points the z elevation was as expected and the 0m point into the cave was at the same level than the first two surface points.
> > > 
> > >             After some tests I realized the start of the cave was the problem.
> > >             You can see this part on the attached picture.
> > >             As you can see there is a vertical start of the cave.
> > > 
> > >             Here the data from that first part:
> > >             data diving station depth newline tape compass
> > >             0   0.0
> > >                         4.9   0
> > >             1   4.9
> > >                         9.1   0
> > >             2   14.0
> > >                       28.1   78
> > >             3   28.2
> > >                         8.9   88
> > >             4   30.3
> > > 
> > >             As soon as I add 0.01m to the first and second (vertical) tape measurement (4.9 = 4.91 and 9.1 = 9.11) all is good and the problem with loopclosure is gone. Both fixed points are added again and the eleveation is as expected (0m level is at the same level at all points).
> > >             So there is some more distance needed for the vertical parts. But I can't explain why.
> > > 
> > >             Is this a problem with the loop closure algorithm?
> > >             May be some of the experts around can post a explanation for this behaviour!?
> > > 
> > >             regards,
> > >             Torsten
> > > 
> > >                 > > > > Bruce Mutton via Therion < therion at speleo.sk mailto:therion at speleo.sk > hat am 30. Oktober 2018 um 08:34 geschrieben:
> > > > 
> > > > 
> > > >                 Torsten
> > > > 
> > > >                 The image of the xvi file makes it look as though there is a loop (not sure if it is a plan or elevation view – presume it is elevation).  Also if you have a fixed point at each end, then this could also cause ‘a loop’ which might give differing elevations at each end.
> > > > 
> > > >                 Also, if the water is flowing, then the water levels at each end will actually differ.
> > > > 
> > > >                  
> > > > 
> > > >                 But I assume you have correctly assessed these effects as negligible, or at least not detected by your survey method.
> > > > 
> > > >                  
> > > > 
> > > >                 I think you are implying that the 0m line is a raster background.  If that is the case, it may be distorted by Therion.  One way to check would be to turn on gridlines for the layout.  That will show you what Therion really thinks is a horizontal line.
> > > > 
> > > >                  
> > > > 
> > > >                 Also check that Therion has the water surface survey stations at the same altitude.  You can check this easily by interrogating a .3d output with Aven.
> > > > 
> > > >                  
> > > > 
> > > >                 Bruce
> > > > 
> > > >                  
> > > > 
> > > >                 From: Therion <therion-bounces at speleo.sk mailto:therion-bounces at speleo.sk > On Behalf Of Torsten Schnitter via Therion
> > > >                 Sent: Tuesday, 30 October 2018 04:36
> > > >                 To: therion at speleo.sk mailto:therion at speleo.sk
> > > >                 Cc: Torsten Schnitter <torsten.schnitter at netcologne.de mailto:torsten.schnitter at netcologne.de >
> > > >                 Subject: [Therion] ***UNCHECKED*** Problem with elevation view
> > > > 
> > > >                  
> > > > 
> > > >                 Hi all
> > > > 
> > > >                 I have a under water cave survey and all the data is done with depth measurements.
> > > >                 So the Z coordinates are absolute and the survey is not with clino data.
> > > >                 The cave starts at one point and after roughly 800m you have another surface point.
> > > > 
> > > >                 When I compile the data I get a strange behaviour.
> > > >                 Although it's not a WYSIWIG map editor I would expect points which have the same depth are at the same "depth"/level on the xvi file. But they are not (see attached screenshot "elevation problem").
> > > >                 The 2 points within the red marks do have the same depth measurements.
> > > >                 But if you have a closer look to it you can see they are not at the same level.
> > > >                 (They differ somehow in x/y way and therefore the points are not on top of each other.)
> > > >                 There are no loop closures inbetween which could be a reason for different levels.
> > > > 
> > > >                 Not having the same level with these points (I assume it is this point) leads to a bigger problem later on.
> > > >                 When drawing the elevation scraps and putting a water surface (depth = 0m) to the scraps and having a raster in the background, the surface at the beginning of the cave is not at the same level than at the next surface point of the cave. But it should be.
> > > >                 See pictures "elevation problem-start" and "elevation problem-end".
> > > > 
> > > >                 What's wrong?
> > > >                 Can anybody help?
> > > >                 If more info is needed please let me know.
> > > > 
> > > >                 Thanks,
> > > >                 Torsten
> > > > 
> > > >             > > > 
> > > 
> > >              
> > > 
> > >                 > > > > _______________________________________________
> > > >                 Therion mailing list
> > > >                 Therion at speleo.sk mailto:Therion at speleo.sk
> > > >                 https://mailman.speleo.sk/listinfo/therion
> > > > 
> > > >             > > > 
> > > 
> > >              
> > > 
> > >             <Entrance-1.JPG>_______________________________________________
> > >             Therion mailing list
> > >             Therion at speleo.sk mailto:Therion at speleo.sk
> > >             https://mailman.speleo.sk/listinfo/therion
> > > 
> > >         > > 
> > 
> >     > 
> 
>      
> 
>         > > _______________________________________________
> >         Therion mailing list
> >         Therion at speleo.sk
> >         https://mailman.speleo.sk/listinfo/therion
> > 
> >     > 
> 
>      
> 


 

> _______________________________________________
>     Therion mailing list
>     Therion at speleo.sk
>     https://mailman.speleo.sk/listinfo/therion
> 


 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.speleo.sk/pipermail/therion/attachments/20181103/0d37adf1/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Cave-Lake-without-loop-closed-plan-view.JPG
Type: image/jpeg
Size: 271336 bytes
Desc: not available
URL: <http://mailman.speleo.sk/pipermail/therion/attachments/20181103/0d37adf1/attachment-0004.jpe>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Cave-Lake-without-loop-closed-elevation-view.JPG
Type: image/jpeg
Size: 356894 bytes
Desc: not available
URL: <http://mailman.speleo.sk/pipermail/therion/attachments/20181103/0d37adf1/attachment-0005.jpe>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Cave-Lake-with-loop-closed-plan-view.JPG
Type: image/jpeg
Size: 107012 bytes
Desc: not available
URL: <http://mailman.speleo.sk/pipermail/therion/attachments/20181103/0d37adf1/attachment-0006.jpe>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Cave-Lake-with-loop-closed-elevation-view.JPG
Type: image/jpeg
Size: 222293 bytes
Desc: not available
URL: <http://mailman.speleo.sk/pipermail/therion/attachments/20181103/0d37adf1/attachment-0007.jpe>


More information about the Therion mailing list