vgwit

  • **
  • 90
Some issues with Next firmware
« on: April 23, 2019, 02:24:37 pm »
Hi,

a week ago have flashed a latest Next code into my 250 quad, erased all settings and configured from scratch. Have too little flights so far but have some questions. All of them are not critical but nevertheless...

One is new - Altitude in GCS's Firmware tab does not correspond to altitude in PFD. Not sure but on 16.09 it was consistent.

Another two questions came from 16.09. Real RTB velocity does not correspond to its value in FlightModeSettings > ReturnToBaseVelocity. Default value of 2 gives about 1 m/s in reality.

And last and probably more complicated. I've mentioned it already some time ago. A way my quad flies to base in RTB mode depends on previous flight mode. In case I switch from AltVario (or any non-GPS mode) to RTB the quad behaves as expected (goes up then slowly to base). But in case of switching from VelocityRoam (or any GPS mode) it does not goes up first but rushes to base much faster (with increasing altitude like on a ramp) and passes base point by couple of meters. Here is a footage:

https://drive.google.com/open?id=1jiAI4SBINMby_NdnYtsZvq9oqfEcfWan

Bank settings are almost identical.

Re: Some issues with Next firmware
« Reply #1 on: April 24, 2019, 04:12:27 am »
I remember the video you made of the RTB speed issue.  I don't know if it has been fixed in next.

Quote
Altitude in GCS's Firmware tab does not correspond to altitude in PFD. Not sure but on 16.09 it was consistent.
I wouldn't have expected that.  Are you configured the same way in next?  You may need to be using INS13 to get correct fusion between baro and GPS?

f5soh

  • *****
  • 4560
    • LibrePilot
Re: Some issues with Next firmware
« Reply #2 on: April 24, 2019, 12:02:21 pm »
Quote
Altitude in GCS's Firmware tab does not correspond to altitude in PFD. Not sure but on 16.09 it was consistent.
Behaviour is the same with 16.09.
PFD: relative altitude above Home location
deluxebaro / Fw tab: Raw baro altitude = uncalibrated baro altimeter

https://bitbucket.org/librepilot/librepilot/pull-requests/293/lp-367-set-return-to-home-velocity/diff

Re: Some issues with Next firmware
« Reply #3 on: April 24, 2019, 07:34:43 pm »
That link shows the fix was merged into 16.09.  I think the issue being described is broken in next but OK in 16.09?  So it is a new issue or regression?

vgwit

  • **
  • 90
Re: Some issues with Next firmware
« Reply #4 on: April 25, 2019, 08:20:56 am »
Behaviour is the same with 16.09.
PFD: relative altitude above Home location
deluxebaro / Fw tab: Raw baro altitude = uncalibrated baro altimeter

OK, probably I'am wrong about 16.09.

The problem with RTH velocity looks the same on 16.09 and Next. I could not find any difference. The same abnormal fast ramp flight and lower than configured velocity in case of correct flight.

vgwit

  • **
  • 90
Re: Some issues with Next firmware
« Reply #5 on: May 02, 2019, 11:16:58 am »
Have an update on RTB behaviour.
In case I switch from any GPS mode to RTB - get a abnormal fast start with ramp flight.
In case I switch the receiver off being in GPS mode - get a correct behaviour. Failsafe is of FC type.

f5soh

  • *****
  • 4560
    • LibrePilot
Re: Some issues with Next firmware
« Reply #6 on: May 02, 2019, 06:13:40 pm »
The behaviour you describe with a fast RTB (with HorizontalVelMax limitation) and ramp is the expected behaviour, there is no sequence (first altitude and next move to Home) for RTB.

Try setting the ReturnToBaseVelocity to 0 and go to failsafe : what happens ?
 

vgwit

  • **
  • 90
Re: Some issues with Next firmware
« Reply #7 on: May 02, 2019, 09:20:32 pm »
is the expected behaviour, there is no sequence (first altitude and next move to Home) for RTB.

Now it is clear that the two flight patterns are determined by two different settings.
And it is good that failsafe flight pattern does not depend on flight mode just before failsafe. But I don't see any logic why there are two different flight patterns for two switch mode transitions (GPS-mode -> RTB and non-GPS -> RTB) without failsafe?