Position Control - Flight Mode

I need some clarifications to better understand this flight mode,

please correct what I miss understand,

  • Position Control is like DRIFT mode in Arducopter. Flying with a gps mode but quite aggressively.
  • Releasing the stickt, will break the copter in the air and it change the mode to HOLD (LOITER)
  • All Parameters in the Multicopter Position Control (MPC) tab, affects the Position Control Mode.

What I expect of this flight mode is, sporty flying with gps “security” and when releasing the stick a perfect steady copter in the air (almost no movement).

At the moment it look like a crazy horse when I release the sticks. In both modes HOLD and Position Control

My thoughts about this problems,
maybe the copter is constantly overshooting and try to get the perfect spot. Is it possible to make the radius larger? So it dose not have to be corrected if he stays in this radius.

I played around with those values, but not big success. It was a bit calmer but not what I need because it is still constantly correcting his position.

  • MPC_ACC_HOR_MAX = 5.0
  • MPC_XY_P = 1.0
  • MPC_XY_VEL_P = 0.9

I hope I’m clear enough otherwise let me know.

1 Like

It’s more like ArduCopter LOITER mode. PX4 doesn’t have a multicopter coordinated turn mode like DRIFT.

Ok… But it is still acting like a wild horse if I switch in position control and not touching the sticks, like in the video.

What setings are needed to calm the copter? How dose it know where to stay after releasing the sticks? What parameter is it?

What is HOLD then?

HOLD is LOITER without any stick input. Do you think your quad is tuned properly? How’s your GPS?

That’s what I try to find out. So in Stabilized it flies like a charm. Perfect hover and even aggressive flying is mastering properly. Altitude controle is okay, in my opinion it could holt the hight better, but I imaging this could be a tuning matter.

In Position control I had about 10-13 Satellites and an hdop of 0.8. So this should not be a problem.

My knowledge about position control and all the settings is quite limited and that’s why I opened this thread because I need and want to understand it.

Here is the log of the flight in the video
http://logs.uaventure.com/view/T82HpEE4irBvrVWbFgH4VC

It looks like you just need to tune the position controller.
https://pixhawk.org/firmware/parameters#multicopter_position_control

@LorenzMeier this issue is still open right? You said a betta release will fix it, right? I didn’t fly since last week so I’m waiting for your new version.

Hi Guys I have experienced the same issue today on a machine.
Flying 1.4.1 Stable version that I took from the standard version in QGC.
Flies perfectly in Stabilized mode, Altitude mode but when I switch to Position Mode, it starts doing wierd thing and flies all over the place. Not stable in position.
I did few things, swapped the Pixracer, the GPS, cables, unplugged arming switch, buzzer, etc. Nothing has improved there. What is weird is that we have built 2x identical machines, the other one is OK. The only difference is that the other machine is using 4x different motors.
If it can help, I never had this issue using 1.4.1 RC1, I will try it again tomorrow if it’s not raining :-/
Regards,
pturcotte

Hi Guys,

Here is the LOG FILE
http://logs.uaventure.com/view/vut88w2Hpop2QPuPD8HWpK

Using Stable 1.4.1
QGC MAC 3.0.0

Here is a video of it:
https://vimeo.com/178623199

Good news!

We are generating charts for the log file titled ‘POSITION FLIGHT MODE ISSUE V1.4.1 F450’.

Plots: http://logs.uaventure.com/view/vut88w2Hpop2QPuPD8HWpK
Log: http://logs.uaventure.com/logmuncher/charts/vut88w2Hpop2QPuPD8HWpK/logmuncher.log

If you would like to upload another log file you can do that now at logs.uaventure.com

Regards,
LogMuncher

Hi Guys,

Did a lot of testing today.

at the end I came back to V1.4.1RC1

This is the one that I have used with QGC

Got it work correctly, much better than V1.4.1 STABLE

Looks like something has changed from 1.4.1rc1 to V1.4.1 STABLE, that’s all I can explain.

I know it is not a hardware issue because I tried it with Arducopter 3.4 RC2 and it was flying well.

Thanks for the support,

Regards,

pturcotte

Hi

The first thing to try is to reduce the vibrations as they are very high. Try balancing your propellers at least.

You can see the vibration levels in this olot: http://logs.uaventure.com/view/vut88w2Hpop2QPuPD8HWpK#IMU_Acc_PLOT

Perhaps you can compare that plot with the one from your other 450.

Simon

Hi Simon,

Props are balanced, motors, not sure…

I will check this for sure. What I don’t understand is when I use 1.4.1RC it works, when I am using Arducopter 3.4RC2 it works. That’s why I am not sure.

I will pay attention to vibration level indeed.

Thanks for the tip !!!

Regards,

pturcotte

It’s not a vibes issue, otherwise it would be very bad in Stabilized as well.
@LorenzMeier wrote that’s a software problem and a fix will be released soon. Hope I didn’t miss understand him.

Just to make it clear, I will not put my pixracer on damping pads. In my opinion this is not necessary :slight_smile:

@LorenzMeier any news about position control? Did you managed to fix it?

I also came across this problem. the position flight mode is not stable

@Sn0west @xuefengleng2016 We have no indications there is anything wrong with position control, it rather seems to be a tuning issue on your end. How can we help you with better tuning instructions to get it right?

For what it’s worth, I’m convinced this is a definite problem and should be taken seriously. The behavior shown in the video published above started immediately after updating to v1.4.1 from v1.4.1rc. I waited for v1.4.2 to see if the problem persists, and I’m afraid it does. Recallibration and adjusting tuning sliders does not solve it, but going back to 1.4.1rc does. My craft is an F550 clone with pixhawk hardware.
The symptoms and circumstances pricing the surging behaviour are described perfectly in the postings above.

#describing the surging behavior

@Mark_Whitehorn Would you have time to try to reproduce the issue?

@LorenzMeier I flew 1.4.1 on my AeroQuad with both LPE and EKF2 on about 13 August, and both performed well in POSCTL.
But it’s an X2, not Pixracer.

Is the ICM20608 fix in v1.4.1? I’m not sure how to tell if it’s been back-ported.