No NTUN (anymore?) in the dataflash logs?

[Mihai]

I’m trying to debug a strange behavior we had with a hex-copter (running 4.0.4) - it may be our software, or something else in the arducopter navigation system, and I tried to look for the navigation information in the dataflash log. Documentation says it should be in NTUN group, but there is no NTUN group (anymore)? Also CMD (records the commands given) seems to be missing.

Is there a way to enable them?

Finally, is there a difference between AHR2 and POS Latitude and Longitude? If so, which is what?

Thank you for the help,
Mihai

[dkemxr (Dave)]

@Mihai

but there is no NTUN group

NTUN was replaced by the PSCx fields a long time ago. Type “ntun” in the search box here for a more comprehensive answer.

[amilcarlucas]

Here is a table explaining the LOG_BITMASK bits and how they affect the logged messages:

How to methodically tune (almost) any multicopter using ArduCopter 4.4.x

Message Description Log rate LOG_BITMASK field name bit value ACC IMU accelerometer data SCHED_LOOP_RATE IMU_RAW 19 524288 AHR2 Backup AHRS data 25Hz if ATTITUDE_FAST is set, else 10Hz - - - ATSC Scale factors for attitude controller SCHED_LOOP_RATE / 1 ATTITUDE_FAST and ATTITUDE_MED 1 and 0 3 ATSC Scale factors for attitude controller SCHED_LOOP_RATE / 2 ATTITUDE_FAST 0 2 ATSC Scale factors for attitude controller SCHED_LOOP_RATE / 4 ATTITUDE_MED 0 1 ATSC Scale factors for att…

[dkemxr (Dave)]

You should probably replace “NTUN” in the Table with “Navigation Tuning” . Haven’t seen NTUN anywhere in awhile. V4.2 maybe.

[Mihai]

Thank you so much! I didn’t quite find the PSCE, PSCN, and PSCN (probably because I’m running a relatively old firmware (4.0.4.)), but I did find PSC which had some of the things I was looking for (PSC.PX and PSC.TPX - the position in X and target position in X), and the corresponding in Y. I was surprised to see that the target keeps moving. For example, if I give it a command to go 100m N, the target Y only moves a few meters north, but it keeps moving in front of the drone as the drone goes. I’m not sure why PSC.TPX is not going 100m north and the PSC.PX would move from 0 to 100.

If you have any insight please do share!
Thanks,
Mihai

[amilcarlucas]

It’s called a L1 controller on ArduCopter 4.0. Google for it.

[Mihai]

Ah… cool! I didn’t know. I see that it’s used both for the rover and the plane - I guess that for the copter as well (although I didn’t see the reference). It’s an interesting way to do control, but I guess that the MIT paper shows it to outperform the PID by a large margin.

I guess that the eventual final target coordinates are no longer available in any of the variables in the dataflash log then?

Thanks!
Mihai