Demonstration of why Altitude Limits should be removed

[Meta4]

My recollection from ancient times is that consumer grade units were single frequency devices, had limits on the number and quality of receivers, and generally took a very long time to lock in enough sats to compute position. In contrast, military units were dual frequency, and contained more receivers making them far more accurate. I also seem to recall that manufacturers were required to cripple devices as to altitude and velocity so they couldn't be used to pilot ballistic missiles.

Then on the satellite side SA caused the intentional streaming of errors to further reduce accuracy. So I think there were a multitude of factors in play though I agree that turning off SA was a big step in the right direction

The early single channel receivers were clunky and near impossible to use with even light tree cover.
The evolution of receivers advanced and Garmin brought out several 12 channel models from 1997, well ahead of the switching off of SA in May 2000.
The idea of manufacturers crippling devices is probably a myth as SA alone was more than enough to seriously upset precision .
When SA was switched off it was quite unexpected. It had been talked about as something that might happen in a few years and then one day it was just switched off and GPS suddenly was much more accurate..

 

[DCATALO]

I don't like being restricted, but I know it's for my and others safety. Now I also understand that when limits are imposed they normally favor one over the other, so I don't always agree with the limits settings. But as it's already been pointed out that it's because of the bad apples in the bunch that these limits are even set, when you post videos of your breaking these limits the only thing your doing is helping FAA and drone haters . Your not making flying drones better, you and those that do these things and stuff along these lines and post them online for the world to see are just making sure that the limits are going to get tighter. I have no anger towards you, I just get bewildered by the way drone enthusiasts screw themselves and others. I hope nobody ever causes a crash with a jet/plane because if that ever happens you can guarantee that the limits will be serious and the lives effected will be many, and nobody wants this to become reality...just my 2 cents!

 

[SoundByte]

You were 10,400 feet above mean sea level, but never 10,400 above the ground.

True, but that makes no difference to the point he was trying to make.

 

[sar104]

You are correct that Doppler effect is used to measure velocity, but it is also used in the determination and refinement of satellite to receiver ranging. Calculating GPS satellite distance is far more complex than R=CΔt. If GPS sats were geosynchronous, we might not be having this discussion. The fact they have an orbital velocity introduces a Doppler effect into the satellite transmission that impacts ranging. For a more technical discussion:

Doppler Shift | GEOG 862: GPS and GNSS for Geospatial Professionals

(In fact, the entire GEOG862 lesson set is a fascinating read, but I digress.)

The subject of Doppler in this particular forum has no real bearing on the OP or my assertion that DJI has done its consumer base a disservice by failing to provide in-flight AGL though it has the technical means to do so.

Well this thread has turned into quite a detailed GPS discussion, so I'll reply again, just to make sure any confusion is cleared up. The Doppler shift in the carrier is used to calculate velocity (range rate) not position (range), which is what you originally stated and I took issue with. The fact that phase-locking is also used to improve the ranging solution in high-end GPS units doesn't change the fact that range is not a Doppler measurement.

As for the argument over the vertical accuracy of GPS positioning, I think that the simplest way to consider the question is from the basic geometric perspective. A GPS position is a 3-D solution and in terms of that solution, all three dimensions are equivalent. The variation in accuracy between those dimensions is thus not inherent in the system but rather due to the variation in errors arising from the geometric arrangement of the satellites relative to the receiver which, on average, does not change, but favors horizontal position accuracy.

As a result, as GPS receivers have improved, the accuracy in all three dimensions has improved similarly. In the absence of DGPS, RTK or other enhanced techniques that introduce a fixed known point, the time-averaged relative error in the vertical dimension has generally been found to be around 1.5 x that of the horizontal dimensions.

In the US, WAAS-enabled consumer-grade GPS units such as used in Phantoms now have around 3 - 4 m RMS accuracy in the horizontal plane with a clear sky view, and so we should expect to see 5 - 6 m vertical. That's consistent with GPS altitude data that I've seen. Unfortunately its drift occurs on relatively short timescales (seconds), presumably since the satellite constellation arrangement relative to the receiver changes significantly on that timescale. That is perhaps why barometric altitude, even though relative rather than absolute, is a better solution for stabilizing vertical flight. It's accuracy is unimportant compared to its resolution, which is sub-meter, and its drift rate, which is generally slow and due to weather-related pressure changes.

 

[A.L.E.X]

The height limitations are a big thrill killer IMO. But why should the height limit be the same a hundred miles from any airport. One should be able to drive out to the middle of nowhere and do some cloud surfing. I have always wanted to do this, but got my first drone after the limitation was introduced. And what about size? The spark is tiny. How small does the drone have to be before it's considered harmless? DJI could certainly get smarter about where it's safe to remove the height restriction (aka mountain ranges).

 

[SoundByte]

The height limitations are a big thrill killer IMO. But why should the height limit be the same a hundred miles from any airport. One should be able to drive out to the middle of nowhere and do some cloud surfing. I have always wanted to do this, but got my first drone after the limitation was introduced. And what about size? The spark is tiny. How small does the drone have to be before it's considered harmless? DJI could certainly get smarter about where it's safe to remove the height restriction (aka mountain ranges).

Only DJI knows for sure what their reasons are for the height restriction. All kinds of conjectures have been brought forth, but none that I have seen have been substantiated by an internal DJI document. Everyone has their own take on this subject, and usually spin what scant info there is to suit their own situation. My take is that most for-profit companies are concerned mostly with market, cost, and profit. Very few companies are altruistic without a payoff of some kind concerning any combination of these 3 things.

I too, following my own logic, have my suspicions. One thing I think we can all agree on, is that DJI has access to vast amounts of flight data. Every warranty problem, every fly away where the bird was recovered and sent in. All this data paints a picture for DJI, and we can only guess at what it looks like. It determines what changes to make to there product, not us, not the government and definitely not because it's necessarily the right thing to do. The auto industry is a prime example where analysis determines whether a recall happens or they continue to deal with lawsuits.

 

[Bakersfield Quad]

The height limitations are a big thrill killer IMO. But why should the height limit be the same a hundred miles from any airport. One should be able to drive out to the middle of nowhere and do some cloud surfing. I have always wanted to do this, but got my first drone after the limitation was introduced. And what about size? The spark is tiny. How small does the drone have to be before it's considered harmless? DJI could certainly get smarter about where it's safe to remove the height restriction (aka mountain ranges).

The reason you cant go cloud surfing, is because it is illegal, to protect those that fly in the air. You just don't get it, it is not about you and the fun you could have, it is about rules and regs to protect those in the air. It could be you, your kids, your mother, who cares who it is, you cant fly with the plane! If you want to do that, get a pilots certificate and fly an airplane.

 

[Jav]

Why is it ok to fly at you want except if you have a phantom standard. Limit them all. That's fair

 

[sar104]

Why is it ok to fly at you want except if you have a phantom standard. Limit them all. That's fair

All Phantom models later than the P2 have the same altitude restriction of 500 m, don't they?

 

[Jav]

Do they all moan 2?

 

[skyeboysteve]

A couple years ago I came across an article (Cheap Centimeter-Precision GPS For Cars, Drones, Virtual Reality) about developing refined algorithms that gets GPS accuracy down to centimeters without requiring a huge antenna like survey grade GPS requires.

I've never heard that GPS uses doppler calculations. I think that would be irrelevant as the speeds of the satellites are so much higher than any vehicle that's processing the signals. As others have mentioned triangulation occurs with timing signals corrected for time dilation. Despite the GPS satellites traveling in less curved time-space than us terrestrial dwellers, their relative high velocities (stemming from energy/acceleration [which itself warps local space-time] that they used to go into orbit in the first place) has caused their local time to run slower than ours.

As for using an onboard database with terrain--that's impractical for most of the world (outside the US and some major world cities) as the SRTM resolution is one elevation point per 90m x 90m area.

 

[sar104]

A couple years ago I came across an article (Cheap Centimeter-Precision GPS For Cars, Drones, Virtual Reality) about developing refined algorithms that gets GPS accuracy down to centimeters without requiring a huge antenna like survey grade GPS requires.

I've never heard that GPS uses doppler calculations. I think that would be irrelevant as the speeds of the satellites are so much higher than any vehicle that's processing the signals. As others have mentioned triangulation occurs with timing signals corrected for time dilation. Despite the GPS satellites traveling in less curved time-space than us terrestrial dwellers, their relative high velocities (stemming from energy/acceleration [which itself warps local space-time] that they used to go into orbit in the first place) has caused their local time to run slower than ours.

As for using an onboard database with terrain--that's impractical for most of the world (outside the US and some major world cities) as the SRTM resolution is one elevation point per 90m x 90m area.

GPS uses Doppler for range rate, and thus velocity, simply because it is much more accurate than the first time derivative of the code correlation range data.

 

[sar104]

@Meta4, @DefiantChild - in case you are interested I pulled the GPS and barometric altitude data from a DAT file conversion that explicitly included both. Below is the comparison for one flight lasting around 13 minutes. The first graph is the simple comparison (with the GPS altitude zeroed at the start). Note that the GPS alttitude fluctuations (noise) are much higher frequency and larger than the barometric altitude fluctuations.

The second graph shows the RMS difference through the flight, raw and smoothed.

In terms of the overall data, the average RMS difference is 1.1 m with a standard deviation of 0.5 m and a maximum (unsmoothed) difference of 3.15 m. The agreement between the two is very good and illustrates that, at least for this flight, the GPS altitude, at least in differential mode, is much better than some conventional wisdom suggests, and is also comparable to the expected variations based on relative errors in the vertical and horizontal dimensions. The fluctuations in the GPS data, even when smoothed, do give a clue why barometric altitude is preferred for flight control.

 

[SoundByte]

@Meta4, @DefiantChild - in case you are interested I pulled the GPS and barometric altitude data from a DAT file conversion that explicitly included both. Below is the comparison for one flight lasting around 13 minutes. The first graph is the simple comparison (with the GPS altitude zeroed at the start). Note that the GPS alttitude fluctuations (noise) are much higher frequency and larger than the barometric altitude fluctuations.

The second graph shows the RMS difference through the flight, raw and smoothed.

In terms of the overall data, the average RMS difference is 1.1 m with a standard deviation of 0.5 m and a maximum (unsmoothed) difference of 3.15 m. The agreement between the two is very good and illustrates that, at least for this flight, the GPS altitude, at least in differential mode, is much better than some conventional wisdom suggests, and is also comparable to the expected variations based on relative errors in the vertical and horizontal dimensions. The fluctuations in the GPS data, even when smoothed, do give a clue why barometric altitude is preferred for flight control.

View attachment 87422

View attachment 87423

Gotta say, it's rather refreshing to see a little empirical evidence on this subject. I have no idea why DefiantGuy insisted on burying a simple point of view in a mountain of techno speak.

 

[sar104]

Back to the ORIGINAL question, I do not agree with the manufacturer of a drone "governing" (like your lawnmower) any function of the aircraft.

An airplane can fly as fast as its motor will allow, and as high the engine will aspirate, if controlled to do so by the pilot. But the pilot must comply with the rules for the different types of airspace.

In my car, the speedometer reads up to 120 miles per hour. And I have reached speeds of over 110 mph ( in a controlled environment). But the speed limit on the road is still 70 on the highway, 55 on the 4-lane, and 35 in the neighborhood. It is the operator of the vehicle's responsibility to comply with the laws and regulations.

So it is not the manufacturer's responsibility to limit the functions of the vehicle/craft, but rather the operator's responsibility to operate it within the rules, and deal with the consequences if/when those rules are violated.

Yes, back to that question, and once again the difference is that, unlike Part 61 Pilots (or even Part 107 sUAS remote pilots) or vehicle drivers, the vast majority of recreational pilots have precisely zero training and certification, and a significant number of them demonstrably don't care or know about the rules and guidelines, and/or believe that they are optional and don't apply to them, and think that the risk of getting into trouble is negligible. So that's a poor analogy.

And no, it's not a requirement for the manufacturer to introduce such restrictions, but DJI have apparently concluded that it is in the best interests of the industry to try to minimize reckless flying by engineered controls. Whether they succeed is open to debate, as is whether they chose the right controls. And if enough customers vote by defecting to their competitors then that might change their assessment, but no sign of that happening at present.

 

[sar104]

Gotta say, it's rather refreshing to see a little empirical evidence on this subject. I have no idea why DefiantGuy insisted on burying a simple point of view in a mountain of techno speak.

I may have provoked that by pointing out that basic GPS positioning is not a Doppler shift measurement, but both he and @Meta4 did make some interesting points.

 

[SoundByte]

I may have provoked that by pointing out that basic GPS positioning is not a Doppler shift measurement, but both he and @Meta4 did make some interesting points.

True, I didn't see the inner workings of GPS as being germane to the point being made. Just seemed self aggrandizing.

 

[sar104]

True, I didn't see the inner workings of GPS as being germane to the point being made. Just seemed self aggrandizing.

Well maybe, but perhaps it was also unnecessary for me to correct the original statement, since no one else was likely to have known or cared about the difference.

 

[N017RW]

Back to the ORIGINAL question, I do not agree with the manufacturer of a drone "governing" (like your lawnmower) any function of the aircraft.

An airplane can fly as fast as its motor will allow, and as high the engine will aspirate, if controlled to do so by the pilot. But the pilot must comply with the rules for the different types of airspace.

In my car, the speedometer reads up to 120 miles per hour. And I have reached speeds of over 110 mph ( in a controlled environment). But the speed limit on the road is still 70 on the highway, 55 on the 4-lane, and 35 in the neighborhood. It is the operator of the vehicle's responsibility to comply with the laws and regulations.

So it is not the manufacturer's responsibility to limit the functions of the vehicle/craft, but rather the operator's responsibility to operate it within the rules, and deal with the consequences if/when those rules are violated.

A corporation's responsibilities are to its ownership or BOD and investors. For survival it does what is best for it.
Secondarily, yet intertwined, comes industry and customers.
The actions it takes is in their own best interests. Expecting otherwise is futile. So saying it's not 'their responsibility' to do this or that is naive.
Customers or the marketplace can dictate survival and vote with their wallet.

 

[sar104]

So you are saying that my opinion is wrong? It is a perfect analogy. It is not the item, it is the operator that bears the responsibility.

I argued that your analogy is poor, for the reasons that I stated. You are entitled to any opinion you feel like.