[ntp:questions] Sure Electronics GPS board: Amazing performance. :-)

unruh unruh at wormhole.physics.ubc.ca
Fri Apr 1 06:12:56 UTC 2011

On 2011-03-31, E-Mail Sent to this address will be added to the BlackLists <Null at BlackList.Anitech-Systems.invalid> wrote:
> unruh wrote:
>> My only concern is that the unit places its location 12m
>>  away from its actual location on Googleearth-- ie, either
>>  googleearth is 12 m out in its alignment to lat/long grid
>>  at my house, or the unit is misreporting its position.
>> Note that the scatter has only a standard deviation of
>>  about 1.5 m ( coming from its use of WAAS I suspect),
>>  so it is not that that 12 m is just part of the natural
>>  variation in gps.
>> (Does anyone know of anywhere where the alignment errors
>>   of google earth to the grid is discussed?)
> Maps Guide Adam, Google Employee
> You should *not* be using the imagery displayed in Google Earth
>   for this purpose.
>  If you have a project which requires a certain threshold of accuracy
>   and geometric precision you will need to utilize the appropriate
>   photogrammetric methodologies and tools.
>  The coordinates, elevations, distances, and measurements provided
>    by Google are approximations only.
>   Google makes no claims as to the accuracy of these measurements.

Those are weasel words (apparently google was sued by someone who relied
on googleearth and got injured by walking along a highway instead of a
path). Google tries to make their images accurate. The question I have
is, what is the accuracy of the googleearth maps in their geolocations. 
I am sure it varies. Probably better in Vancouver say than on the
atacama in Chile.

>> A brief 10 min test at a survey marker put the unit about
>>  3 m away from that marker on googleearth-- but that was
>>  at a location 6km from my home.
> If you don't have a clear view of the sky,
>  e.g. only out a window on one side of a building
>  the positions certainly are skewed a bit (in my experence).
> Assuming you have a clear view of the sky in all directions,
>  run something like google.com/search?q=SA+Watch for an
>  extended period of time to determine a _fixed_ position
>  more accurately.

In my home it is out a window in the front of the house, and the
position is 12m in that direction out (if I believe googleearth).
> Off the top of my head (IIRC);
>  In general, for vehicle street / map navigation you need
>   a HDOP < 4 (~<52' / 16m)
>  With a CEP 95% probability, the actual position is within
>   66ft / 20m diameter circle of the position a GPS returns.
>  Without several GPS satellites, plus a SBAS satellite
>    (or some other form of DGPS)
>   I think you are very unlikely to continuously stay under
>    3m accuracy, on short time stationary fixes, or mobile
>    fixes with a consumer GPS.

The CEP (1 sigma) over 15 hrs was about 1.5m
It is claiming it is using DGPS (WAAS?)

><nstb.tc.faa.gov/DisplayArchive.htm> has some information
>  about recent Dilution of Precision measurements and GPS
>  accuracy records, ...

More information about the questions mailing list