Can someone please look at my DAT file to see why my drone wont fly straight

Also I've noticed that when I first turn the drone on the camera does it thing going back then forward and then it goes straight but it moves a lot in the view on my iPad it almost like it goes up slow and down or if I move the drone side to side it slowly goes into the middle after a few seconds. I'm not sure if this is related or not. I can provided a screen capture to show if it helps.
 
I'm not sure if this is related or not. I can provided a screen capture to show if it helps.
It could be, but I am going to say that this is not a gimbal issue. The IMU Positioning error is in this data as well. Its co-incidental possibly, but in all 3 data sets, excluding only one that did not have the error, all of these showed up in the first 80 seconds or so of each one. Below shows the error and the gimbal yaw angles which are within 1-2 degrees of the aircraft center.

1572117689624.png
 
It could be, but I am going to say that this is not a gimbal issue. The IMU Positioning error is in this data as well. Its co-incidental possibly, but in all 3 data sets, excluding only one that did not have the error, all of these showed up in the first 80 seconds or so of each one. Below shows the error and the gimbal yaw angles which are within 1-2 degrees of the aircraft center.

View attachment 115256


Ok I think we're narrowing it down. Do you think the IMU Positioning error is something to be worried about. Also would this error affect the drones flight path?
 
Also I've noticed that when I first turn the drone on the camera does it thing going back then forward and then it goes straight but it moves a lot in the view on my iPad it almost like it goes up slow and down or if I move the drone side to side it slowly goes into the middle after a few seconds. I'm not sure if this is related or not. I can provided a screen capture to show if it helps.

As long as it returns to center each time probably not, as in the video of mine the gimbal did not return to center. It reacted normal otherwise to movement of the drone but true center was off. And it varied from moment to moment how far off it'd be.

What I see you doing in the videos is hovering lining up for flight following the dirt road for example. Then you fly forward with no fine rudder adjustments after initial line up, to actually get it to follow the road.

Odds you'd be dead on and from that high in the air from a stand still are slim.

Line up then start flying forward adjusting rudder until follows the line you want. Then see what you think, and reduce altitude for practice might help too. Couple hundred plus feet in the air adds to the effect.
 
It could be, but I am going to say that this is not a gimbal issue. The IMU Positioning error is in this data as well. Its co-incidental possibly, but in all 3 data sets, excluding only one that did not have the error, all of these showed up in the first 80 seconds or so of each one. Below shows the error and the gimbal yaw angles which are within 1-2 degrees of the aircraft center.

View attachment 115256
Do you know what causes the positioning error? I get that in some of my flights as well.

Many times flying the same location/flight path and with either one of my drones.
Sometimes there but other times not and I get that GPS Position NoMatch more often with my mavic than I do with my P4. Never noticed a problem caused by it, nerve-racking to find it after the fact. It is not posted on the dji go screen during flight.

Oh and like blong I live out in farm country, fly an old rail road bed across the road for testing purposes and now and then just for quick fun.
 
Do you know what causes the positioning error?
Not exactly. That is one reason I asked @sar104 to look at the .dat IMU data. The "GPS Position No Match" message is seen quite often, but does not normally effect the flight. Between him and @BudWalker , they can better answer your questions in that regard. As of note, I have seen this issue with a P4 a few times, but I'm not sure if there was any definative conclusion on the exact cause.
 
Not exactly. That is one reason I asked @sar104 to look at the .dat IMU data. The "GPS Position No Match" message is seen quite often, but does not normally effect the flight. Between him and @BudWalker , they can better answer your questions in that regard. As of note, I have seen this issue with a P4 a few times, but I'm not sure if there was any definative conclusion on the exact cause.
Yea until it came up in conversion I wouldn't have known as seems to suffer no ill effects. I reviewed a few flights and it goes all the way back from when the drones where new.

And yea again, I enjoy reading all of Budwalker's and sar's posts, the both of em definitely are know how in action.
 
That rudder command was intentional I suspect and is not part of the area I was looking at. I am looking between 167-219 seconds for a straight flight with no inputs. You can see the +2 to 3 degree roll with only the elevator applied.

View attachment 115255
I've watched that segment a few times now and just not seeing what you see after the last commanded turn at 147.913 and then the turn at 219.852. Raining today so can't stream in HD, may or may not be a factor, probably not but .

There is a twitch at 228.921 following his rotating around to the left followed by slight and slow correction the the right (4m 10sec mark). The camera twitches slightly "left" just as or maybe just a hair before he lets completely off right rudder.
As to recenter itself But nothing logged I see to account for it in the player and by no means rotating fast enough at that point for synchronized gimbal pan to act?
 
Thank you so much for looking into this. What does this mean in layman's term. Is one of the motors failing or going bad?

No - the motors are fine, but compass 0 appears to be being affected by the adjacent motor power cable. Here is a better view of those data:

mag_motor_current_2.png


It's more common to see the compass data affected by tilt and yaw, which implies an uncalibrated magnetic characteristic on the aircraft, but that doesn't appear to be the case here - it really looks correlated with motor current. That has been seen before, if I remember correctly, due to poorly routed or untwisted motor power cables.

To summarize, since there is so much speculation in this thread, the problem seems to be that the IMU is getting systematically bad heading data from the Mag0. That means that it is not facing the direction that it thinks it is, by up to 10° in some cases. When you apply elevator to move forwards, that results in the actual track not matching the (incorrect) heading, and so it crabs to correct. The FC thinks that the aircraft (and camera) are facing the direction of travel.

The offset is variable in magnitude and direction since it depends on the interaction of the spurious magnetic field (that appears to be motor current induced) with the earth's magnetic field.
 
  • Like
Reactions: BudWalker
No - the motors are fine, but compass 0 appears to be being affected by the adjacent motor power cable. Here is a better view of those data:

View attachment 115264

It's more common to see the compass data affected by tilt and yaw, which implies an uncalibrated magnetic characteristic on the aircraft, but that doesn't appear to be the case here - it really looks correlated with motor current. That has been seen before, if I remember correctly, due to poorly routed or untwisted motor power cables.

To summarize, since there is so much speculation in this thread, the problem seems to be that the IMU is getting systematically bad heading data from the Mag0. That means that it is not facing the direction that it thinks it is, by up to 10° in some cases. When you apply elevator to move forwards, that results in the actual track not matching the (incorrect) heading, and so it crabs to correct. The FC thinks that the aircraft (and camera) are facing the direction of travel.

The offset is variable in magnitude and direction since it depends on the interaction of the spurious magnetic field (that appears to be motor current induced) with the earth's magnetic field.

Well that makes a heck of a lot more sense than "what do you expect from consumer level drone" that I got over there on that other forum. I didn't think it was right either and see the effect in many videos posted on the WWW, distance flights for example.

Appears there is a cause and effect behind it after all and now something solid to work with to hopefully correct it.

One question if I may, where did you get the source of the "Mag_0_Mod_smth" plot from in the signals list? I looked under Mag(0) but can't seem to match it.
 
Well that makes a heck of a lot more sense than "what do you expect from consumer level drone" that I got over there on that other forum. I didn't think it was right either and see the effect in many videos posted on the WWW, distance flights for example.

Appears there is a cause and effect behind it after all and now something solid to work with to hopefully correct it.

One question if I may, where did you get the source of the "Mag_0_Mod_smth" plot from in the signals list? I looked under Mag(0) but can't seem to match it.

Mag_0_Mod_smth is processed - it doesn't exist in the DAT file or the DatCon conversion. It's Mag_0_Mod with a rolling 50 point binomial smoothing function applied to make the data more easily visualizable.
 
Mag_0_Mod_smth is processed - it doesn't exist in the DAT file or the DatCon conversion. It's Mag_0_Mod with a rolling 50 point binomial smoothing function applied to make the data more easily visualizable.


That explains it and also why I mostly just follow along!

I see it in both of mine, sometimes it's as it should be but sometimes no not so much.

Interacting with you guys in this thread? I know that it is not a normal condtion for these drones, possible cause and confident that it is a correctable issue going forward.
 

Members online

No members online now.

Forum statistics

Threads
143,066
Messages
1,467,358
Members
104,935
Latest member
Pauos31