FAA sUAS 400’ AGL Requirement

[sar104]

I'd think they'd implement it so that just the immediate area around the flight is downloaded/used...and that should make the file sizes quite manageable.

And as far as the cost goes, they look to be free in the U.S. from the USGS...

Where can I get elevation data?

Download free digital elevation data through The National Map Download Client. Click the “How To” link at the top of the viewer for detailed instructions on how to find and download data; see The National Map training videos for step-by-step demonstrations.Download standard elevation products...

www.usgs.gov

How often would you suspect people are taking off from locations that are at a significantly different elevation than ground level? I've heard of people taking off from the roof of a building, or maybe the balcony of a hotel at the beach, but that seems quite rare to me. You could always compare the GPS data to the DEM value for the take-off location and ask the user for confirmation if there's a discrepancy.

Honestly, I'm not at all familiar with DEMs and how accurate the data is. I just guessed that there might be a way to download only a portion of the data, similar to the way an Android user can download portions of the Google map data, without the need to store hundreds of GB of information. And I would store all of the data and do all of the calculations on the users device, rather than in the aircraft or the RC.

But then again, I am only looking at this as a (far from expert) semi-pro programmer. And things always seem to get more complicated once you start writing code.

The problem with downloading DEM data as needed is that it relies on a reasonable internet connection. But that would be a possibility if the lack of height AGL in many situations were acceptable.

With 30 m DEM resolution the takeoff elevation is going to be significantly wrong in many situations - and typically worst in the kind of terrain where height AGL would be most important. And my guess is that DJI won't do this, if only because most flights are at heights where the ground elevation is only one part of the equation, i.e. the height of structures and vegetation is significant compared to the aircraft height. I foresee many people simply trusting the height AGL, flying into trees and building, and then blaming DJI.

 

[okw]

The problem with downloading DEM data as needed is that it relies on a reasonable internet connection. But that would be a possibility if the lack of height AGL in many situations were acceptable.

With 30 m DEM resolution the takeoff elevation is going to be significantly wrong in many situations - and typically worst in the kind of terrain where height AGL would be most important. And my guess is that DJI won't do this, if only because most flights are at heights where the ground elevation is only one part of the equation, i.e. the height of structures and vegetation is significant compared to the aircraft height. I foresee many people simply trusting the height AGL, flying into trees and building, and then blaming DJI.

Again, I would download the data the same way I download the Google Maps data (if that's possible) ... at home when connected to Wi-Fi. Internet connection would not be a problem.

The link that wsalopek posted above to the USGS data shows resolutions down to 1 meter. But I say again, I don't know anything about DEM data, so maybe I'm misunderstanding what's available.

I completely agree that it is highly unlikely that DJI will ever include this functionality to its app. Just saying that it's possible to do without adding anything to the aircraft.

 

[GadgetGuy]

Maybe this: SF11/C (120 m)

It's relatively easy to make it work with equipment such as Pixhawk flight controllers.
If your DJI equipment (Matrice 210 or M600) is capable of interfacing PWM or IC2 signals, you are good to go.

That laser only works to 40m over water, rather than 120m. Perhaps the existing 10m ultrasonic downward sensor could be enhanced, by adding a laser. Not perfect, but it might work!

 

[Drgnfli]

I think I kinda understand what you all are saying.
It’s just seems that if I can get relatively accurate altitude readings from my iPhone & a map app while hiking, it wouldn’t be that difficult to install software for the drone to relay altitude at AGL. I do admit I’m very technologically challenged !!!

Yes, don't all the GPS apps use satellites to give me my altitude in my car? I've been thinking about that since I got my MP. Wait a second, I'll bet the GPS location already has the altitude, doesn't it? Darn! Thought I had something there...?

 

[superstarpup]

Yes, don't all the GPS apps use satellites to give me my altitude in my car? I've been thinking about that since I got my MP. Wait a second, I'll bet the GPS location already has the altitude, doesn't it? Darn! Thought I had something there...?

GPS altitude is inaccurate, significantly so when discussing 400’ limits.

Sar, how does the barometric altimeter in mavics work? Can the altimeter be set?

Realistically, this is just an academic discussion as it’s all just a wag anyways. anything that will be accurate to more than 75’ or so will likely be far too bulky and heavy for a small drone.

 

[DanMan32]

Even with barometric AGL relative to takeoff, I've sometimes found significant drift.
I encountered a 7' differential within minutes. I was surprised that VPS wasn't providing more accurate reading.

What's funny at least now is that I went to reset height by landing. Problem was said landing was at the far end of the field and forgot the landing also reset home point.
I don't recall it telling me home point was updated, but whatever.

 

[Meta4]

Even with barometric AGL relative to takeoff, I've sometimes found significant drift.
I encountered a 7' differential within minutes. I was surprised that VPS wasn't providing more accurate reading.

It's surprising that you think VPS wasn't accurate.
VPS gives a separate height reading from the barometer and it's usually very accurate.

What's funny at least now is that I went to reset height by landing. Problem was said landing was at the far end of the field and forgot the landing also reset home point.
I don't recall it telling me home point was updated, but whatever.

Did you turn the drone off and on again?
It's motor start up that sets the home point, not landing.

 

[sar104]

GPS altitude is inaccurate, significantly so when discussing 400’ limits.

Sar, how does the barometric altimeter in mavics work? Can the altimeter be set?

Realistically, this is just an academic discussion as it’s all just a wag anyways. anything that will be accurate to more than 75’ or so will likely be far too bulky and heavy for a small drone.

The barometric altitude is calculated from the change in ambient pressure from the pressure at take off, based on a standard atmosphere. It has sub-meter resolution when combined with IMU data, and accuracy that varies with atmospheric conditions - usually within 10% of actual altitude change. It cannot be set - you would have to manually add the takeoff elevation and subtract the ground elevation to get height AGL.

Note that GPS altitude doesn't suffer from the standard atmosphere assumption, and while it has less resolution its accuracy is not altitude change dependent. For flights that climb over 100 m or so it's generally more accurate than the barometric altitude.

 

[sar104]

Even with barometric AGL relative to takeoff, I've sometimes found significant drift.
I encountered a 7' differential within minutes. I was surprised that VPS wasn't providing more accurate reading.

What's funny at least now is that I went to reset height by landing. Problem was said landing was at the far end of the field and forgot the landing also reset home point.
I don't recall it telling me home point was updated, but whatever.

VPS should be pretty accurate - provided that the aircraft is flying over reasonably well-defined ground. Over trees, for example, it will not read the distance to ground, and so is not going to give you an accurate height AGL.