Michal Podhradsky

@kirkscheper would limiting the I gain slider to 512 fix the problem? Or should we use 64bit int for the calculation to avoid the overflows altogether?

Is this with Vectornav?

@gautierhattenberger good idea, that would be the next step.

@gautierhattenberger I am trying to figure out if this is a problem on the GCS side, or the AP side. What kind of coordinates does the GCS show? Lat/lon, ECEF...

Not yet, it is in the pipeline - waiting for better weather to flight test.

So what do you think is a good solution? Should I add logic as if `INS_FINV_USE_UTM` is defined, or the other case? I would be adding logic into [ins_vectornav_init()](https://github.com/paparazzi/paparazzi/blob/master/sw/airborne/subsystems/ins/ins_vectornav.c#L163) function

@karthikbhargavan thanks for the hint, please have a look at my new PRs

@msprotz thanks!

Hi @beurdouche sorry I missed the notification here. It works as expected, just tested it. Is there a reason this new `kremlin.h/c` is not yet in the main repository? (here...

OK, I ll post the unused variables warnings once the new kremlib is up.