Hi All,
Just performed a small experiment indoors (it will most probably do exactly the same outdoors), where i tested what the downward facing cameras actually do, besides gound avoidance.
I positioned my Mini on my coffee table, and after takeoff i let the drone settle down into a hover over the table. Once whatever GPS it was receiving calibrated, the mini's position was absolutely stationary... Lovely...
I then began to drag the coffee table around the floor slowly, a little bit, just 1 - 2 feet in various directions, stopping inbetween.  Interestingly the Mini followed the table where i dragged it and repositioned itself over the centre of the table!

I was mildly impressed..
Anyway there ya go.
Don't try this at home folks, or do.. it's up to you.. :-D
Phased.

Yeah it is within VPS range so it achieves more precise hover by referencing what it sees below. This does not work if your mini is 10m+ above your tea table.

Guorium Posted at 7-16 05:20
Yeah it is within VPS range so it achieves more precise hover by referencing what it sees below. This does not work if your mini is 10m+ above your tea table.

LOL, if i was 10m above my coffee table, I'd be in a bedroom upstairs!

You might like to compare flying it over a completely stationary surface and then over one where the sensors can see something flapping in the down draught, I am fairly sure mine responds to the flaps. In the latter situation mine is not as staionary as it is in the former but the fact that it seems to respond to flapping really impressed me.

I'd guess that there can be downsides to this and maybe that's why I have had the odd rejected landing but I still think it impressive

cool test!

But this is exactly what is happening hovering low above floating water....

GPS wants to keep it in postion, while the vision will move the craft with the streaming water. A problem for the software....

cheers
JJB

Interesting test - thanks for sharing your findings

Once whatever GPS it was receiving calibrated, the mini's position was absolutely stationary...
It's got nothing to do with GPS.
It's all the downward facing sensors and will do the same with no GPS.

Sounds like a limited version of Active Track.  Some people should be very happy.

Labroides Posted at 7-16 06:35
Once whatever GPS it was receiving calibrated, the mini's position was absolutely stationary...
It's got nothing to do with GPS.
It's all the downward facing sensors and will do the same with no GPS.

Even when over a uniform surface with no features to latch onto?

PhasedSpaces Posted at 7-16 09:16
Even when over a uniform surface with no features to latch onto?

Even when over a uniform surface with no features to latch onto?
Obviously the ability of the downward facing sensors is subject to the limitations that are listed in your manual.
But given suitable conditions, they provide perfect stability without any GPS.

When you fly at night or poor light the drone drifts even with gps so it's the sensors keeping it steady at low flight levels. Also flying indoors I doubt you have any gps, the drone is using the sensors for stability and perhaps explains why it follows the table pattern. If you did the same test outside i'd be interested to see how far the drone would follow the table before it locked back on gps coordinates.

Hi. Thank you for creating this informative post and sharing your experience with your DJI Mavic Mini. This information will be a great help to our valued DJI co pilots who owns a DJI Mavic Mini. Thank you for your valued support and safe flying.