Welcome, Autel Pilots!
Join our free Autel drone community today!
Join Us

Orbital Rotation Question

RJ_Make

Well-Known Member
Joined
Aug 30, 2016
Messages
245
Reaction score
54
I'll be the first to admit that I'm still very much honing my control skills, but I seem to be having a problem controlling the aircraft rotation smoothness. I can do a fairly controlled orbital rotation with my P3S, but on my XSP, I seem to be having an issue with the left RC stick input.

The issue is even with the most even, gradual left or right stick input it will not ramp the spin. Once it gets to a certain point it's 'full on', then I slowly let go and it's 'full stop'.

I can't find any setting to dampen the stick input, but maybe someone else know where to look?
 
  • Like
Reactions: Tpat591
Funny, I just added this to the "wish list" on that separate thread. I've also found it very difficult to control yaw smoothly and accurately for horizontal panning in order to get good videography shots with the XSP, even after the "fix" provided by the latest firmware update (1.2.8) that theoretically "improved the throttle curve for yaw to be more linear". I'm not sure what that means, but it doesn't seem to have had the result I was hoping to see. I would really like to see a way for the user to at least adjust the yaw sensitivity for the control stick, if not all of the flight axis inputs. Currently, there is no way to do that.
 
  • Like
Reactions: Tpat591
I have spoken to Autel about incorporating expo adjustment in the next firmware. They said they would look into it. A controller without expo adjustment is not common. Every controller I have ever used, even cheap ones have expo adjustment. Hope they listen to users more.
 
One way to do smooth pans is to use Orbit. Click on Orbit. Confirm the center is the aircraft (the default), then "Start". Optionally set the radius, direction, and # of laps. The radius defaults to 49ft but can be set down to 33ft, which will result in a faster orbit/pan. I have started setting the "Laps" to 3-5 and just exiting if I'm ready to quit before it finishes. I often use a different gimbal angle for each rotation.
 
I did try that over the holiday weekend, TomK, with the hopes that the autopilot controlled orbit would do a better job than my ham-handed thumbs for panning, but I really got similar results both ways. On the autopilot orbit, it looks like the ship made a yaw correction to face the target once every few seconds. In other words, it would orbit for a few seconds, then change yaw while continuing to orbit, orbit for a few more seconds, then change yaw again while orbiting, etc. Or maybe it is continuously yawing slightly while orbiting, and every few seconds making a larger yaw correction. Regardless,the overall result was a fairly "choppy" looking pan - not worse than I was doing manually - and in some respects better - but definitely not what I would consider smooth. I'm on the road right now, but I'll try posting a video example of my orbit pan once I get home.

I hate to keep looking like I'm complaining more than emphasizing the good points of the XSP! I really am quite happy with the ship overall. I just want to a) see if I can change what I'm doing to get more out of the ship and b) make sure I'm not expecting too much. All my previous drone experience is with small microdrones - I have nothing to adequately compare my experience with.
 
I have spoken to Autel about incorporating expo adjustment in the next firmware. They said they would look into it. A controller without expo adjustment is not common. Every controller I have ever used, even cheap ones have expo adjustment. Hope they listen to users more.
Glad to hear they will at least be considering it for future updates. I'll mention it to CS as well - the more folks they hear from who think it's an important feature, the more likely it will be to happen.

Also, thanks for giving me the proper vocabulary for the control stick sensitivity - I've read a bit more on expo adjustments in general now. If I'm interpreting what I've read correctly, then the fix introduced in the new firmware may have actually worsened the yaw from a "smoothness" standpoint? Autel's firmware download page says that release 1.2.8 "improved the throttle curve for yaw to be more linear". In order to get a slower, smoother pan with initial stick inputs for yaw, wouldn't I want the yaw curve to be more exponential? Or am I misunderstanding how expo works?

Edit: I guess that I'm assuming they already had "positive" expo on the RC controller to begin with, and then made it more linear. Is it possible that Autel had the RC controller set with "negative" expo with the previous firmware, such that making "the yaw curve more linear" really has helped? Negative expo would seem to be a bad idea for a ship targeted as an aerial photography platform, but I guess it's a matter of taste.

I thought a saw a little improvement in yaw with the firmware update, but the firmware came very soon after I bought. The improvement was limited enough that it could just as easily be chalked up to me getting better/more comfortable flying the XSP.

Quagmire, if you see this thread - do you still have one XSP on the old firmware and the other running the new firmware? If so, do you see a difference in yaw control between the two?
 
Last edited:
Glad to hear they will at least be considering it for future updates. I'll mention it to CS as well - the more folks they hear from who think it's an important feature, the more likely it will be to happen.

Also, thanks for giving me the proper vocabulary for the control stick sensitivity - I've read a bit more on expo adjustments in general now. If I'm interpreting what I've read correctly, then the fix introduced in the new firmware may have actually worsened the yaw from a "smoothness" standpoint? Autel's firmware download page says that release 1.2.8 "improved the throttle curve for yaw to be more linear". In order to get a slower, smoother pan with initial stick inputs for yaw, wouldn't I want the yaw curve to be more exponential? Or am I misunderstanding how expo works?

Edit: I guess that I'm assuming they already had "positive" expo on the RC controller to begin with, and then made it more linear. Is it possible that Autel had the RC controller set with "negative" expo with the previous firmware, such that making "the yaw curve more linear" really has helped? Negative expo would seem to be a bad idea for a ship targeted as an aerial photography platform, but I guess it's a matter of taste.

I thought a saw a little improvement in yaw with the firmware update, but the firmware came very soon after I bought. The improvement was limited enough that it could just as easily be chalked up to me getting better/more comfortable flying the XSP.

Quagmire, if you see this thread - do you still have one XSP on the old firmware and the other running the new firmware? If so, do you see a difference in yaw control between the two?


You are correct in the way you interpret exponential and linear. It is a matter of taste, that's why it should be adjustable. As far as your choppy panning, are you shooting 30fps? If you shot 60fps you panning will be butter smooth. I have never gotten smooth panning with any camera at 30fps. Also try keep the shutter speed to twice the frame rate. This is what is called the 180 shutter. This with will give you smooth moves and the right motion blur to make it look natural. If you are not concerned with motion blur and want a bit more sharper look then go up to 360 shutter, or 4 times frame rate. Hope this helps.
 
Last edited:
I did try that over the holiday weekend, TomK, with the hopes that the autopilot controlled orbit would do a better job than my ham-handed thumbs for panning, but I really got similar results both ways. On the autopilot orbit, it looks like the ship made a yaw correction to face the target once every few seconds. In other words, it would orbit for a few seconds, then change yaw while continuing to orbit, orbit for a few more seconds, then change yaw again while orbiting, etc. Or maybe it is continuously yawing slightly while orbiting, and every few seconds making a larger yaw correction. Regardless,the overall result was a fairly "choppy" looking pan - not worse than I was doing manually - and in some respects better - but definitely not what I would consider smooth. I'm on the road right now, but I'll try posting a video example of my orbit pan once I get home.

I agree that the rotation is not constant during the Orbit, but it is smoother than I can do manually, or at least consistently do manually. Changing the Orbit radius should have an effect. I was thinking that increasing the radius would smooth it out, but it might make the yaw correction less frequently which would make it less consistent. I usually use a radius of 33-35ft, close to the minimum, and it seems fairly smooth. Not constant, but smooth.
 
  • Like
Reactions: Glazier
You are correct in the way you interpret exponential and linear. It is a matter of taste, that's why it should be adjustable. As far as your choppy panning, are you shooting 30fps? If you shot 60fps you panning will be butter smooth. I have never gotten smooth panning with any camera at 30fps. Also try keep the shutter speed to twice the frame rate. This is what is called the 180 shutter. This with will give you smooth moves and the right motion blur to make it look natural. If you are not concerned with motion blur and want a bit more sharper look then go up to 360 shutter, or 4 times frame rate. Hope this helps.

Thanks for this, X-Starman! Videography is very new to me - all of my previous experience is with still photos - so this is a fantastic tip. Even if it doesn't completely solve the look of my panning specifically, it will certainly help the quality of my videos in general. I'm pretty sure I'm set for 60fps, but I'm positive my shutter speed is set way too high for this rule of thumb - I can definitely see that causing the yaw adjustments to look more pronounced and/or less natural. I'll double check all those settings and experiment on my next flight, probably this weekend. Huge thanks again!
 
I agree that the rotation is not constant during the Orbit, but it is smoother than I can do manually, or at least consistently do manually. Changing the Orbit radius should have an effect. I was thinking that increasing the radius would smooth it out, but it might make the yaw correction less frequently which would make it less consistent. I usually use a radius of 33-35ft, close to the minimum, and it seems fairly smooth. Not constant, but smooth.
Yes, I agree, Orbit is definitely more consistent than what I can achieve manually - not necessarily constant rate, but reliably centered on target and certainly never over-steers like I have done many times trying to micromanage the yaw manually and ruining the shot. It definitely makes sense that radius would have an impact as well. When I tried it over the weekend, it was a middle-of-the-road distance (maybe about 150'). I think the maximum is in the neighborhood of 300 or 400'? Anyway, definitely gives me food for thought and some ideas to experiment with - it'll only make my filming better! Thanks much for all the insight, guys.
 
I did some experiments with different radii/radia/radiuses. The movement along the horizon looks fairly constant to me. I pulled the log file for this flight and will be able to plot yaw v time and delta yaw v time, though I'm not sure when I'll get to it.
 
  • Like
Reactions: RJ_Make
Thanks for posting up that video. It might just be my eyes, but it seem a bit smoother at the larger radius's, buy yeah over all pretty consistent. On another note, what you see @~10:11 is exactly what I get. It's starts off with some lag (may just be perceived and not actual lag), and then I get that pixelation. In a second or 2 it washes over the entire screen, then clears up. I'm now thinking it's interference.
 
I've also heard Autel is aware of the problem and have stated they're working on a fix.
 
Please forgive the double post. I put this reply in another thread, then saw this thread which seems more on topic for my issue.

Same problem here......yaw speed is not constant. When the craft is rotated either clockwise or counterclockwise and the control stick is held in a fixed position, the aircraft yaws fast, then slow, fast, then slow, over and over. I have been in touch w/Autel and they acknowledge that there is a problem and the techs in China are working on it. No ETA on a fix. I'm on my second bird now and if it isn't fixed soon, I will have to return this one for a refund. This is a deal-breaker for me. Too bad, because otherwise, I'm happy with the unit.
 

Latest threads

Members online

No members online now.

Forum statistics

Threads
11,227
Messages
102,646
Members
9,818
Latest member
redwingaerials