Altitude settings in waypoints

[virtue]

As I understand it, the P3 measures altitudes above homepoint (using barometric pressure). Thus with a waypoint recorded at, say, 100 feet altitude the P3 would know to fly to that x-y coordinate point at an altitude of 100' above the start point.

However, if your homepoint for the start of a waypoint flight was LOWER than on the 'training' run, the actual height at x-y would also be lower. If you started 101 feet (vertical distance) down a hill, the P3 would try to plough into the ground at x-y.

Do I have that right?

Which begs the question, why does DJI use barometric pressure to altitude? GPS provides three dimensional positioning. This is absolute positioning, not relative to a starting point. The x-y precision is obviously extremely good; I assume its altitude reading would be similarly accurate. At the very least, even if there is some additional benefit from a barometric measure, why not use both?

 

[N017RW]

GPS is not reliable for altitude.

If you look into it a bit you'll find this out as well as many handheld GPS units using barometers for altitude.

 

[III% Streve]

GPS is not reliable for altitude.

If you look into it a bit you'll find this out as well as many handheld GPS units using barometers for altitude.

+1
It used to be, before phones had barometers in them, that GPS navigation while driving was TERRIBLE while driving in the hills or mountains. I remember looking for a dive shop new Beaver Lake (Yes, I was beaver diving!) and GPS said I was there. The store was more than a mile down the road. Altitude variations totally screw up GPS cords!!

 

[N017RW]

Not sure I can agree with the last sentence.

The accuracy of the 'cords' you refer to has to do with the mountains obscuring SVs not that you are at any given altitude.

Explain why there are no such 'cord' issues while ascending or descending in flight?

 

[RoyVa]

Guess what commercial airplanes use for altitude, it's not GPS ....
Barometric altimeters that are calibrated before take off would be a good guess.

 

[RoyVa]

I set my way points for the altitude I want. If I set the first one it kinda flows to the others. If I set it for two hundred feet and the ground climbs up as with a hill so does my Phantom. So it must take into account altitude above ground level or the AGL altitude. Have you guys seen the same?

 

[virtue]

GPS is not reliable for altitude.

If you look into it a bit you'll find this out as well as many handheld GPS units using barometers for altitude.

Following these comments about GPS accuracy, I looked into it as suggested. There are reasons having to do with the distributions of the positions of GPS satellites with respect to the receiver that reduces altitude accuracy (vs. horizontal accuracy), by a factor of 1.5 to 3 as far as I can tell. However, the P3 is remarkably accurate horizontally, presumably due to its ability to average signals over so many satellites so it is not clear to me that the altitude measure would all that bad.

The barometric solution is only useful if you only need relative altitude measures (or can apply corrections for changing atmospheric conditions, day to day or location to location, which airplanes do). For the most part that is true for drones, but my post was motivated by DJI's release of waypoint capability, for which an absolute vertical setting would important any time the waypoints are used from a different starting point (i.e. at a different elevation). At the very least users should be aware of the altitude errors expected in such situations.

Why not integrate both barometric and GPS data - the P3 has access to both after all?

 

[tcope]

Following these comments about GPS accuracy, I looked into it as suggested. There are reasons having to do with the distributions of the positions of GPS satellites with respect to the receiver that reduces altitude accuracy (vs. horizontal accuracy), by a factor of 1.5 to 3 as far as I can tell. However, the P3 is remarkably accurate horizontally, presumably due to its ability to average signals over so many satellites so it is not clear to me that the altitude measure would all that bad.

The barometric solution is only useful if you only need relative altitude measures (or can apply corrections for changing atmospheric conditions, day to day or location to location, which airplanes do). For the most part that is true for drones, but my post was motivated by DJI's release of waypoint capability, for which an absolute vertical setting would important any time the waypoints are used from a different starting point (i.e. at a different elevation). At the very least users should be aware of the altitude errors expected in such situations.

Why not integrate both barometric and GPS data - the P3 has access to both after all?

If it used both, what would it do when they were different? Take one over the other (you turn don't need two) or average them and always be incorrect?

 

[N017RW]

Following these comments about GPS accuracy, I looked into it as suggested. There are reasons having to do with the distributions of the positions of GPS satellites with respect to the receiver that reduces altitude accuracy (vs. horizontal accuracy), by a factor of 1.5 to 3 as far as I can tell. However, the P3 is remarkably accurate horizontally, presumably due to its ability to average signals over so many satellites so it is not clear to me that the altitude measure would all that bad.

The barometric solution is only useful if you only need relative altitude measures (or can apply corrections for changing atmospheric conditions, day to day or location to location, which airplanes do). For the most part that is true for drones, but my post was motivated by DJI's release of waypoint capability, for which an absolute vertical setting would important any time the waypoints are used from a different starting point (i.e. at a different elevation). At the very least users should be aware of the altitude errors expected in such situations.

Why not integrate both barometric and GPS data - the P3 has access to both after all?

I'm just pointing out how it works.

It's not remarkable.
Anyone using the same u-blox GPS engine can achieve the same results.

 

[virtue]

If it used both, what would it do when they were different? Take one over the other (you turn don't need two) or average them and always be incorrect?

Just one possibility - in waypoints, if the start position altitude (GPS) was different from the start altitude when the waypoints were defined, notify the user and ask if all waypoint heights should be adjusted accordingly.

 

[tcope]

Just one possibility - in waypoints, if the start position altitude (GPS) was different from the start altitude when the waypoints were defined, notify the user and ask if all waypoint heights should be adjusted accordingly.

But you are right back to where you started...

You take off at Ground level A. So this is now 0.

You then change to Ground level B which is 10' lower.

The Phantom has two figures... GPS report's 0 and Barometer reports -10'. So... according to what you posted... no warning. Your still assuming GPS is always the correct reading.

I agree with you that saving the waypoint flight and taking off from a different altitude _is_ going to cause an issue. I can see this being an issue if it's something like 50' or so as the pilot should be allowing at least 50' of clearance if not more.