Re: [Paparazzi-devel] Rotocraft: Increase in psi while switching into HOVER_Z_HOLD mode

[Felix Ruess]

Hi Simon,

it looks all correct to me...

If you plot them together you can see that it correctly uses the current heading when switching to HOVER:

Cheers, Felix

On Tue, Apr 28, 2015 at 10:37 PM, Simon Liebold <address@hidden> wrote:

Hi Felix,

I uploaded the logs here.

Not sure if I plotted (or logged) correctly and/or enough data. Here is another screenshot showing est_psi and sp_psi. I certainly did yaw the copter before and after switching into hover mode - but not while or shortly after switching.

I am noticing that my sp_psi and sp_psi show a noticeable difference. Something I need to work on.

What I don't understand is that guidance_h_heading_sp (sent as cmd_heading in HOVER_LOOP) changes even when you switch from KILL to HOVER.

You meant the switch from KILL to ATTITUDE_DIRECT (not HOVER), or? At time 231.401 (the first "drop" of cmd_heading):

230.207 2 ROTORCRAFT_STATUS 0 0 2 0 3 0 0 0 0 0 125 298
230.806 2 HOVER_LOOP 0 0 423 35 4851 -13654 7 -69 0 0 0 0 0 0 0 0 0
230.864 2 HOVER_LOOP 423 34 423 35 4803 -15526 -5 -56 0 -1 -3 -2 0 0 -6 15 -7190
231.401 2 ROTORCRAFT_STATUS 0 0 0 11 3 4 0 0 2 1 125 300

Mhm, not sure. I guess that is normal. I certainly didn't alter the code in this area (or any other).



All the best,
Simon

Am 28.04.2015 um 21:03 schrieb Felix Ruess:

Hi Simon,

when entering hover, the heading RC setpoint is set to current the heading in guidance_h_hover_enter:
https://github.com/paparazzi/paparazzi/blob/master/sw/airborne/firmwares/rotorcraft/guidance/guidance_h.c#L534

As long as you don't change the heading via RC, that initial heading when entering hover is applied as cmd_heading in guidance_h_run:
https://github.com/paparazzi/paparazzi/blob/master/sw/airborne/firmwares/rotorcraft/guidance/guidance_h.c#L363

So it would make sense to plot the estimated heading together with cmd_heading (e.g. psi from ROTORCRAFT_FP).

My guess is that your (estimated) heading drifted between switches....

What I don't understand is that guidance_h_heading_sp (sent as cmd_heading in HOVER_LOOP) changes even when you switch from KILL to HOVER.

But the only two instances where this variable is updated (apart from initialization to zero) is in guidance_h_run if you are in HOVER or NAV mode.

Did you make any changes that could impact this?

I can take a look at the logs if you upload them somewhere...

Cheers, Felix

On Tue, Apr 28, 2015 at 6:33 PM, Simon Liebold <address@hidden> wrote:

Hello all,

During last flight session I noticed the quad turning clockwise every time I switched into hover mode. The logs seem to support my observation (see screenshot).

Anyone who noticed something like this before? I was expecting HOVER_Z_HOLD to keep the heading when entering this mode.

Simon

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel