Invalid WKT CRS in Annex B - GEOPOSE LOCAL FRAME OF REFERENCE SPECIFICATIONS

[rouault]

Examples at https://docs.ogc.org/is/21-056r11/21-056r11.html#toc47 are invalid WKT.

A valid WKT CRS cannot just start with BASEGEOGCRS. I assume here the top node should be a PROJCRS (not totally sure. I'm hesitating with a GEODCRS[], in which case CONVERSION should be DERIVINGCONVERSION). There was also a missing closing bracket to end DATUM.

A more valid example would be:

PROJCRS["WGS 84 / LTP-ENU",
    BASEGEOGCRS["WGS 84",
        DATUM["World Geodetic System 1984",
            ELLIPSOID["WGS 84",6378137,298.257223563]],
        PRIMEM["Greenwich",0,
            ANGLEUNIT["degree",0.0174532925199433]],
        ID["EPSG",4979]],
    CONVERSION["To LTP-ENU",
        METHOD["Geographic/topocentric conversions",
            ID["EPSG",9837]],
        PARAMETER["Latitude of topocentric origin",<latitude>,
            ANGLEUNIT["degree",0.0174532925199433],
            ID["EPSG",8834]],
        PARAMETER["Longitude of topocentric origin",<longitude>,
            ANGLEUNIT["degree",0.0174532925199433],
            ID["EPSG",8835]],
        PARAMETER["Ellipsoidal height of topocentric origin",<height>,
            LENGTHUNIT["metre",1],
            ID["EPSG",8836]]],
    CS[Cartesian,3],
        AXIS["topocentric East (U)",east,
            ORDER[1],
            LENGTHUNIT["metre",1]],
        AXIS["topocentric North (V)",north,
            ORDER[2],
            LENGTHUNIT["metre",1]],
        AXIS["topocentric height (W)",up,
            ORDER[3],
            LENGTHUNIT["metre",1]],
    USAGE[
        SCOPE["unknown"],
        AREA["To be specified"],
        BBOX[-90,-180,90,180]]]

[rouault]

not totally sure. I'm hesitating with a GEODCRS[], in which case CONVERSION should be DERIVINGCONVERSION)

question asked to http://lists.ogc.org/pipermail/coordtran.wg/2024-June/002767.html

[rouault]

Very comprehensive answer from Roger Lott in http://lists.ogc.org/pipermail/coordtran.wg/2024-June/002768.html that I would sum up along the lines "Presenting a topocentric CRS either as derived geodetic CRS or a projected CRS is fine, with the later being probably more practical for most software implementation"

[3DXScape]

@rouault Thank you. I agree with your analysis. Your corrected example should replace the current one.
Bottom line: this is an example and should be accepted by specialists, e.g, you and Roger Lott. Please comment to confirm that you are happy with your suggested WKT as written above and we will make the replacement in the document.

[rouault]

Please comment to confirm that you are happy with your suggested WKT

yes I am

[3DXScape]

Document update per comments above. Will close when new version appears on .OGC link

[rouault]

Document update per comments above

There are errors. See my comments in https://github.com/opengeospatial/GeoPose/commit/804f8284f7b0dbbae236e7c1067743ce766a9c1a. (I'd suggest using pull requests instead of direct commits)