I've just been experimenting with the Geolocation tools in AutoCAD 2015 and I've noticed a bit of an issue with the data here in the UK.
Its common practice to treat the origin of the World UCS as the origin of the Ordnance Survey national grid.
The Geolocation dialog gives me the latitude and longitude of the dropped marker. So using http://gridreferencefinder.com/ this can be converted to the OS National Grid X,Y.
E.g. 50.2641,-5.1090 converts to 178520,045084 (178520000,045084000 as architects here tend to work in millimeters).
Placing the geolocation marker at World UCS coordinates 178520000,045084000 in a mm file should locate the live map in the correct place for us to overlay the digital mapping data obtained from the Ordnance Survey.
Some experiments in the Bristol area (in the regaion of 51.44, -2.59) I found the livemap data was about 5m too far west and 1m too far south. Easily fixed by tweaking the marker position. I've tried the different coordinate systems that seem to relate to the UK and make no difference.
However, when testing on a location in Cornwall using the coordinates above the data was not misplaced by much but the rotation was well off. Actually nearly, but not exactly, by the amount of magnetic variation in that location (approx 2.6 degrees). Correcting this roation showed the OS map lining up reasonably well with the live map. But this didn't solve my placement issues in the Bristol area.
Its not clear from the documentation how the live map handles north and how, if at all, the application of a GIS Coordinate system should correct magnetic to true or grid north.
Can anyboby shed some light on this?