How to deal with LRUD data?

[M.J. Green]

Hi,

As Phil says, a normal vector is required to make sense of LRUD data. 
Cross-sections needed to be centred around a point, but this could be part 
way along a leg. Although how often these mid-survey leg cross-section have 
well defined centres I do not know.

Many people use a convention to define the normal vector eg. take LRUD data at 90 degrees to a leg, or at the bisector of two legs. If someone uses a consistent style they want to have a convenient way of entering it. Forcing people to use a particular convention is probably not a good idea, although easier to implement. This could probably be done in the same way that different sorts of survey leg can be entered. 

As LRUD data are basically splay shots, we could consider them the same as splay shots, and NSEW data in chambers, etc.  Although Splay shots may be measured more accurately, which may mean that some people may wish to distinguish them.

The interface could be something like:

Syntax
*cross <style> <ordering>

Description <style> = 
VNORMAL|NORMAL|VBISECTOR|BISECTOR|VBEARING|HNSEW|HSPLAY|SPLAY|MIDLEG|MIDLEGSPLAY|HMIDLEGSPLAY|GUESS

<ordering> = ordered list of instruments - which are valid depends on the 
style.


VNORMAL
For vertical planes defined perpendicular in the horizontal plane to a leg
CrossSectionStation OtherStation (Direction) Left Right Up Down 
;Direction options are F (Forward) and B (Backwards), with F being default
1 2 F 1.4 1.2 0.1 0.5
1 2 F 1.4 1.2 0.1 0.5

NORMAL
For planes defined perpendicular to a leg - perhaps nobody does this
CrossSectionStation OtherStation (Direction) Left Right Up Down
1 2 towards 1.4 1.2 0.1 0.5
1 2 away 1.4 1.2 0.1 0.5

VBISECTOR 
For vertical planes defined by the bisector of two legs in the 
horizontal plane 
FromStation CrossSectionStation ToStation Left Right Up Down 

BISECTOR
For planes defined by the bisector of two legs - perhaps nobody does this
FromStation CrossSectionStation ToStation Left Right Up Down
1 2 3 1.4 1.2 0.1 0.5

VBEARING
For vertical planes defined by a compass direction
CrossSectionStation Bearing Left Right Up Down
1 324 1.4 1.2 0.1 0.5

HNSEW
For horizontal planes
CrossSectionStation  North South East West NE SE SW NE

HSPLAY
For horizontal splay shots
CrossSectionStation Compass Tape (Direction)

SPLAY
For splay shots
CrossSectionStation Compass Clino Tape (Direction)

MIDLEG 
For vertical planes defined perpendicular in the horizontal plane to 
a leg, centred somewhere on a centreline 
FirstStation SecoundStation Proportion|Distance (Direction) Left Right Up Down ;Direction options are F 
(Forward) and B (Backwards), with F being default 1 2 F 1.4 1.2 0.1 0.5 1 2 
F 1.4 1.2 0.1 0.5

MIDLEGSPLAY
For mid splay shots
FirstStation SecoundStation Proportion|Distance Compass Clino Tape (Direction)

HMIDLEGSPLAY
For horizontal mid splay shots
FirstStation SecoundStation Proportion|Distance Compass Tape (Direction)

GUESS 
For when we do not know the plane the LRUD data was entered. Uses 
some sort of guess, either perpendicular to the first leg it was included 
in or the bisector of the first two. Probably the direction would point in 
the direction of increasing numbers. i.e. taking LRUD data in the direction 
in which you are surveying. 
CrossSectionStation Left Right Up Down

It is missing the ability the choose a weighting between various legs, i.e. similar to BISECTOR, but more arbitary, perhaps that is rather akward to actually use.

Points on arbitrary cross-sections, could be entered as a series of splay 
shots. Let me know if the list above misses any way that cross-section 
dimensions are commonly recorded.

Cheers,
Martin

On Feb 14 2005, Philip Balister wrote:

> You are touching on a couple of issues with LRUD data:
> 
> 1) LRUD is a station attribute, not a shot attribute. So the LRUD data
> is unique to the station. Survex already treats shots and stations
> separately, so it is in good shape to handle LRUD without the extra shot
> hacks needed by some other cave survey packages.
> 
> 2) Once you have the LRUD data associated with the station, you face the
> problem of what direction the vectors point. In normal cases, LRUD is
> roughly normal to the passage bearing. However, at bends and junctions
> LRUD become almost meaningless. To be consistently useful for map
> drawing, the LRUD data should include a normal vector that defines the
> plane the data was taken in.
> 
> 3) I you really badly need the LRUD data to draw a feature, go ahead and
> take splay shots :)
> 
> Philip
> 
> On Mon, 2005-02-14 at 11:05 -0200, Carlos Henrique Grohmann wrote:
> > In my opinion, LRUD data should be dealt a little different than it is 
> > by most software. You see, create a 3D view of the cave is nice, 
> > beautifull, but that's not the most important thing. Most people draw 
> > their maps in CAD or design software (like CorelDraw, Illustrator), so 
> > the basic need is a centerline with indications of left-right in each 
> > station for plan view (or up-down for profiles). I'm not a good 
> > programmer (just started to lern C), but I don't thonk that's to hard 
> > to do, it could be a 'pseudo-leg' with bearing 90 degrees from the 
> > survey shot itself. One thing I notice about most software, is that 
> > they always assume that LRUD belong to the 'from' station. that's not 
> > so good, because you always have to create false shots to be sure all 
> > stations have their LRUD data, like this:
> > 
> > from  to  tape  compass  clino  L  R  U   D
> > 1     2   10    45       2      2  1  0.5 1   (LRUD of station 1)
> > 2     3   24    56       -1     1  0  4   1   (LRUD of station 2)
> > 2     4   23    67       1      -  -  -   - 
> > 2     5   12    180      1      -  -  -   - 
> > 2     6   20    13       6      -  -  -   - 
> > 4     4   -      -        -     1  2  0   0.5 (LRUD of station 4)
> > 5     5   -      -        -     2  1  2   0   (LRUD of station 5)
> > 6     7   4     89       -1     1  0  4   1   (LRUD of station 6)
> > ...
> > 
> > It would be nice if the program could understant from which station 
> > LRUD data belongs to, like this:
> > 
> > from to tape compass clino station L R U D 1 2 10 45 2 1 2 1 0.5 1 
> > (LRUD of station 1) 2 3 24 56 -1 2 1 0 4 1 (LRUD of station 2) 2 4 23 
> > 67 1 4 1 2 0 0.5 (LRUD of station 4) 2 5 12 180 1 5 2 1 2 0 (LRUD of 
> > station 5) 2 6 20 13 6 6 1 0 4 1 (LRUD of station 6) 6 7 4 89 -1 
> > ........
> > 
> > This is a very simple way to solve the problem, and make easier to 
> > input data. (note: this is how most brazillian speleo groups collect 
> > survey data).
> > 
> > I would like id Survex had something like this. I have tried Therion, 
> > but I couldn't get it to work well. I'm sorry, but in my opinion, the 
> > Therion Book dows not explain well how to work with LRUD data and how 
> > to export centerline.
> > 
> > That's it, just my two cents.
> > 
> > Best regards
> > 
> > Carlos Grohmann - Guano
> > GPME - www.gpme.org.br
> > 
> > 
> > -- 
> > +----------------------------------------------------------+
> >           Carlos Henrique Grohmann - Guano  
> >    Geologist M.Sc - Doctorate Student at IGc-USP - Brazil
> >         Linux User #89721  - guano at usp dot br
> > +----------------------------------------------------------+
> > 
> > 
> > Citando Mike McCombe <mikemccombe at btinternet.com>:
> > 
> > > Carlos, Ten years ago, I wrote "Speleogen", a post-processor to add 
> > > LRUD data to a Survex centreline, creating a multi-layer DXF. It used 
> > > to be on the Survex "downloads" page, though it was Windows-only. In 
> > > the intervening years, I upgraded and re-wrote Speleogen in Java but 
> > > have never released it as in-built LRUD support in Survex has always 
> > > been "coming soon". Although I've only used it on Microsoft and 
> > > Symbian, I'm pretty confident it would also work on Linux and other 
> > > Java platforms.
> > > 
> > > I'd be happy to hear from anyone who'd like to try it out, play with 
> > > the source-code or just debate how LRUD data should be handled!
> > > 
> > > All the best,
> > > Mike 
> > > 
> > > 
> > > -----Original Message----- From: survex-bounces at survex.com 
> > > [mailto:survex-bounces at survex.com] On Behalf Of Carlos Henrique 
> > > Grohmann Sent: 09 February 2005 19:14 To: survex at survex.com Subject: 
> > > LRUD
> > > 
> > > Hello all,
> > > 
> > > so, since survex can't deal with LRUD data, how to get a centerline 
> > > with left-right dimensions in it working on linux?
> > > -- 
> > > +----------------------------------------------------------+
> > >           Carlos Henrique Grohmann - Guano  
> > >    Geologist M.Sc - Doctorate Student at IGc-USP - Brazil
> > >         Linux User #89721  - guano at usp dot br
> > > +----------------------------------------------------------+
> > > 
> > > 
> > > 
> > > 
> > > 
> > > -- 
> > > Survex http://lists.survex.com/mailman/listinfo/survex
> > > 
> > > 
> > > 
> > 
> > 
> 
>