What happens if you turn off VPS -and- positioning at first motor start?

[Droning on and on...]

What happens? I tested it for all of you, and made a video.

Conditions: VPS and OA are disabled via the switch in ••• –> safety –> advanced. Then, when taking off for the first time after startup, when the prompt shows to fly in Normal mode vs. disable positioning, select disable positioning.

What it appears actually occurs: Positioning using the very accurate downward cameras is disabled, and the barometer is also ignored for holding altitude.

Surprisingly, position hold is NOT disabled. It's absolutely clear the FC is still responding with roll/yaw movements to try and hold position. In the video you can see significant gusts and breeziness looking at foliage, while at the same time the drone is twitching around and doing a pretty decent job of staying mostly in place... and here's the key: (Not) coincidentally, within about a 10' circle, a.k.a. GPS accuracy.

The behavior looks very much like GPS is being used for position hold, and GPS altitude is being used for altitude hold.

Have a look:

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

 

[Meta4]

and the barometer is also ignored for holding altitude. ...
and GPS altitude is being used for altitude hold.

What makes you think this?
DJI drones don't use GPS for altitude

Surprisingly, position hold is NOT disabled.

That shouldn't be surprising, the specs all have something like this:
Hovering Accuracy Range
Vertical:
±0.1 m (with vision positioning)
±0.5 m (with GNSS positioning)

Horizontal:
±0.3 m (with vision positioning)
±0.5 m (with high-precision positioning system)

 

[Droning on and on...]

What makes you think this?
DJI drones don't use GPS for altitude

The fact that it doesn't hold altitude. That's my explanation. What's yours?

 

[Kilrah]

The fact that it doesn't hold altitude.

It does though, just not very precisely due to turbulent air. There is no circumstance or mode on consumer DJI drones (other than the FPV/Avata) where there is no altitude hold, and it's always based on baro if there's no VPS.
Altitude hold is more precise when VPS is on when you're close, but what you see there is what happens when you're out of VPS range.

 

[Droning on and on...]

What makes you think this?
DJI drones don't use GPS for altitude

Because it doesn't hold altitude, drifting upward as much as 20ft (7m) and then having to be corrected. You can see this occur in the video over about 10ft (3m) and then I correct with some control input.

That shouldn't be surprising, the specs all have something like this:

Hovering Accuracy Range
Vertical:
±0.1 m (with vision positioning)
±0.5 m (with GNSS positioning)

Horizontal:
±0.3 m (with vision positioning)
±0.5 m (with high-precision positioning system)

Surprising because of this prompt:



I interpret "No Position Hold Mode" to mean there is no position hold. I'm led to believe that even more by implication because there is a separate control in the settings to explicitly disable VPS, not position hold.

How do you interpret the text of the dialog above, and what effect do you expect?

In any case, the vertical and horizontal accuracy of both vertical and horizontal position hold is far outside the spec you posted with this mode selected ("no position hold"). How do you explain this?

I've given mine.

 

[Droning on and on...]

Okay, ran a test.

There are 3 possible sources of altitude information in the Mini 4P: Barometer, IR sensors, and GPS. For this test, the VPS cameras and IR 3D sensors were covered, eliminating them as a source.

The video clearly shows a very big difference between selecting "Normal" (position hold) mode, and "No Position Hold mode".

In normal mode, while there is horizontal drift as expected (recall that VPS/OA is disabled and the sensors are covered), it also clearly is exhibiting Position Hold behavior but with a larger error. This is as expected if using GPS for horizontal positioning.

Altitude Hold is rock solid. No more than a few inches, at most, variation. Since the bottom sensors are covered, the only sources of altitude are barometer, and GPS.

I conclude the FC is using the barometer for altitude data. GPS simply is not accurate enough to hold altitude within inches for minutes at a time.

The second flight with the only difference being starting the motors with "No Position Hold mode" drifts horizontally about the same, but altitude is drifting as much as 15 feet.

My explanation for the difference between the two flights is the barometer is also ignored in the second case, and GPS altitude is being used. Or possibly no altitude hold at all, but I have drones without altitude hold, and they're much more volatile vertically than this gentle drift.

To be clear, I don't care about being "right". I want to know what's right. These are some new features/settings on the Mini 4P, so I tested them, and am trying to explain the behavior. @Kilrah and @Meta4, if you can offer a different theory that explains the behavior, I want to discuss it.

Posting specs that the observed behavior doesn't follow doesn't explain the behavior.

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

 

[MS Coast]

To be clear, I don't care about being "right". I want to know what's right. These are some new features/settings on the Mini 4P, so I tested them, and am trying to explain the behavior. @Kilrah and @Meta4, if you can offer a different theory that explains the behavior, I want to discuss it.

Good on you for raising the question and trying to sort out the meaning of DJI's new normal/disable position hold setting. (As discussed in other threads, there's no apparent difference in the two modes for normal flight settings with VPS and OA enabled.)

My greatest disappointment with DJI is their failure to document new features and operating modes. We should not have to guess about the function of an operating mode choice we're required to make each time we fly.

 

[Droning on and on...]

My greatest disappointment with DJI is their failure to document new features and operating modes. We should not have to guess about the function of an operating mode choice we're required to make each time we fly.

Yes indeed. I have a word for the documentation people, but because this is a family site, I'll only say it begins with "w", ends with "rs", and has an "anke" in the middle