Possibly wrong unit for high-precision accuracy values

[YaguangZhang]

Possibly wrong unit for high-precision accuracy values

It seems the unit for the high-precision accuracy values (outputs from methods getHorizontalAccuracy and getVerticalAccuracy defined in src/SparkFun_Ublox_Arduino_Library.cpp) should have a unit of 0.1 mm, instead of mm as indicated in the example examples/ZED-F9P/Example9_GetHighPrecisionGeodeticPacket. The ZED-F9P Interface Description mentions that there is a scaling of 0.1 for the high-precision accuracy values in Section 5.14.6.1 High Precision Geodetic Position Solution.

Workbench

We are following the the tutorial GPS-RTK2 Hookup Guide using the recommended hardware:

• SparkFun GPS-RTK2 Board
• SparkFun BlackBoard

Steps to reproduce

One can observe an inconsistency of the accuracy values, for instance, by running the examples ZED-F9P/Example8_GetHighPrecisionPositionAndAccuracy and ZED-F9P/Example1_GetPositionAccuracy, and comparing the output values from getHorizontalAccuracy (the 2D horizontal accuracy) and getPositionAccuracy (3D accuracy).

Expected behavior

One should expect the 2D horizontal accuracy to be smaller than the 3D one, because the latter considers the vertical error, too.

Actual behavior

However, if the high-precision accuracy values have a unit of mm as the 3D accuracy does, the 2D accuracy would be way bigger. For example, by outputting both accuracy values to the serial monitor, one can observe something like:

where the 2D accuracy (~290) is way bigger than the 3D one (~50).

[PaulZC]

Hi @YaguangZhang,
We have added a fix for this to the library: #90
Can you please download and test the new code and check that it corrects the issues you were seeing?
ZED-F9P\Example8_GetHighPrecisionPositionAndAccuracy shows how to use the new functions.
I am closing this issue. Please reopen it if you are still having problems.
Best wishes,
Paul