Meta4
Premium Pilot
- Joined
- Jan 8, 2014
- Messages
- 15,303
- Reaction score
- 9,247
Unfortunately WAAS is only available in North America.
The rest of the world has to do without.
Trustworthiness of "home" feature
- Thread starter denemante
- Start date
The rest of the world has to do without.
Meta4
Premium Pilot
- Joined
- Jan 8, 2014
- Messages
- 15,303
- Reaction score
- 9,247
That's pretty much what I was saying in an earlier post. On RTH, mine will usually land about 3 meters or about 10 feet from the take off point. Sometimes it is closer than that, sometimes a little further. It will never be exact with the GPS systems we use.
Please do not confuse two different issues ...
The error or difference in mapping to a GPS position is based on the method of position determination for the maps ... many are still based on very old surveys done by Royal Navy and so on hundred or more years ago.
The earth is also an Oblate Spheroid causing errors in location.
On charts - once GPS surveys are carried out in the area covered - a note will be added to advise the correction to be applied to accurately plot on the map.
I think you will also find that most 'flex' / distortion checking is by lasers and similar equipment.
Accuracy of GPS even military parallel system is not pinpoint. Its better than the civilian spec - but still has its circle of determination.
Nigel
- Joined
- May 28, 2016
- Messages
- 83
- Reaction score
- 20
- Age
- 59
While i might have fallen out of love with my P3s, i have to say the GPS and especially the RTH has always impressed me, i had a total signal fail (my fault) after flying it 500m max out over the village, signal got wiped out by interference and i panic`ed for a couple of seconds, then RTH kicked in and she just flew back and landed, sweet, as said make sure home point is recorded
Meta4
Premium Pilot
- Joined
- Jan 8, 2014
- Messages
- 15,303
- Reaction score
- 9,247
It's often said but isn't really important.
The home point is automatically recorded as soon as the GPS gets a good position fix, which is usually before anyone launcjhes anyway.
But if you did launch early, the GPS is going to get a fix within a few seconds as it flies away and you still would have a home point very close to where you are.
SA was switched off by Presidential Decree. SA was imposed onto the Civilian GPS signal to reduce its accuracy in the mistaken view that it would stop its use in terrorism etc. SA became a joke as Ground Stations were setup by many countries to send a correction signal to later designed Receivers ... this then brought the position calculations back to non SA accuracy ... 2 - 5m ...
WAAS was an addition that FAA introduced which replaced DGPS ... the Ground Station correction. In fact WAAS is not only augmentation system ... Europe .. Japan ... and possibly India have their versions - all compatible with GPS sat signals. But it still only brings accuracy to 3m or less.
Some think that having GLONASS and / or GALILEO will give better accuracy - actually not quite. What it does is increase the probability of more accurate position determination through the receivers averaging of positions. When you get away from the 'hikers' style of GPS Rx ... you find you can increase or decrease averaging time - this is to alter the apparent accuracy of position dependent on where in the world you are. Most people will leave as default.
Don't forget that just because you have 18 ... 19 ... or whatever sats locked on - it does not mean all of them are providing for your position. The software that takes the data - filters out those that fall outside of mean .. and then averages out the retained. The number of sats used in a 'fix' is not fixed and is variable dependent on that mean.
Of course some are quite happy to believe marketing hype and so-called accuracy specs quoted - but in reality - they are hard to replicate.
Just for interest - I sailed on vessels that were testing Magnavox Transit SatNav systems - UGH !! Then later with GPS / Glonass / EGNOS .... much better ! We used both Civilian and Military for 'ops' .....
I'll leave it to Garmin - a respected GPS manufacturer - the last words :
Loading site please wait...
Nigel
- Joined
- Jul 29, 2016
- Messages
- 446
- Reaction score
- 235
- Age
- 81
There appears to be a confusion about how to think about GPS accuracy. The base civilian GPS that DJI uses has a positional accuracy of +- 16 feet 95 % of the time. The chances of hitting the center of that circle once it has been recorded at takeoff is low. The drone is continuously recalculating its position, and each recalculation is giving a slightly different position, even if the drone doesn't move.
If you want to have. 95% chance of the drone landing free of hazards, set the home point in the middle of a 16 foot radius circle. My P3 nearly always is within 8 feet of HP, but not always. No amount of modification of antennas or anything else can change this. The accuracy is defined by the crystal clocks in the satellites.
If you want to have. 95% chance of the drone landing free of hazards, set the home point in the middle of a 16 foot radius circle. My P3 nearly always is within 8 feet of HP, but not always. No amount of modification of antennas or anything else can change this. The accuracy is defined by the crystal clocks in the satellites.
Davis and Nigel are both correct. Placing two recievers side by side does not give an indication of the GPS accuracy. In fact the atmospheric phase correction, one of the largest error sources, is identical in both recievers. If you came back an hour later you might read a error of several meters. This is in fact the basis of differential GPS. That is what WAAS is. You have a reciever at a known location and find the error between the GPS calculated position and the known position. This error is then transmitted to the GPS receiver you are using and corrects your position. Post #19 is confusing differential GPS with the normal, unaided system. The Jet Propulsion Laboratory in Pasadena, CA has a differential GPS system covering southern California for seismic studies which can measure mm changes in the many stations.
A lon-lat system is defined which is tied to the solid Earth. That is the system a GPS receiver reports in. The absolute accuracy of a position measurement is about 5 meters. When you launch your drone the stored “home” location is only as good as this 5 meters. When you return home it is a completely new measurement of position, again with a 5-meter accuracy. Therefore you are unlikely to land exactly where you took off. All this has nothing to do with maps. The maps were originally drawn using celestial navigation and surveying techniques. Since we can now measure absolute position on the Earth to a few meters we find that some of the maps and charts have large errors, up to 10 km for some islands. These are gradually being corrected. When you observe a difference between your map and GPS it is the map that is wrong.
As a highly controversial foot note, some have suggested that during the cold war both US and Soviet maps had large errors artificially introduced so that GPS guided ICBMs would miss their intended targets by a few km.
N017RW
Premium Pilot
The Cold War period ended about 4 years before GPS was fully operational. (1991 vs 1995 respectively)
Sat nav was in design stages in the late 50's and early 60's as navigation for missile systems such as Polaris. This was the reason for Transit ....
Transit accuracy was so variable though due to 'bunching of SATs' and which part of the world you were in. Our testing for Magnavox ... we would have good fixes spread out but as days passed the fixes would bunch up and we would be without fix for hours .... not good in Palawan Passage where we were operating !
The next generation GPS was developed to correct the innacuracies of Transit - giving good coverage over 80% of the world. As you approach polar regions it reduces ....
It was found to be accurate enough to place a guided weapon within a few meters of target. With the development of the parallel civilian system - it was decided to introduce random errors to avoid its use by terrorists - Selective Availability. Various commercial company's developed Differential units to counter this artificial error. Finally SA was switched off as it was not providing the desired results and thwarted by DGPS.
When I worked with Compagnie Generale de Geophysique of France . We had because of their Military connection - military as well as civilian GPS. It was still not accurate enough to conduct scientific seismic work.CGG in fact had the most accurate system - Syledis. But that's another story.
Differential stations were discontinued when SA was switched off. Civilian GPS then attained similar accuracy to the military system of around that 2 - 3 m average.
With regard to our P units .... I have many times landed within 1m of Home point ... just because accuracy wanders - does not mean it cannot achieve a good spot landing. But lets be honest - most will cancel the landing part of RTH and manually land to avoid tip-over and be on good spot.
Nigel
Transit accuracy was so variable though due to 'bunching of SATs' and which part of the world you were in. Our testing for Magnavox ... we would have good fixes spread out but as days passed the fixes would bunch up and we would be without fix for hours .... not good in Palawan Passage where we were operating !
The next generation GPS was developed to correct the innacuracies of Transit - giving good coverage over 80% of the world. As you approach polar regions it reduces ....
It was found to be accurate enough to place a guided weapon within a few meters of target. With the development of the parallel civilian system - it was decided to introduce random errors to avoid its use by terrorists - Selective Availability. Various commercial company's developed Differential units to counter this artificial error. Finally SA was switched off as it was not providing the desired results and thwarted by DGPS.
When I worked with Compagnie Generale de Geophysique of France . We had because of their Military connection - military as well as civilian GPS. It was still not accurate enough to conduct scientific seismic work.CGG in fact had the most accurate system - Syledis. But that's another story.
Differential stations were discontinued when SA was switched off. Civilian GPS then attained similar accuracy to the military system of around that 2 - 3 m average.
With regard to our P units .... I have many times landed within 1m of Home point ... just because accuracy wanders - does not mean it cannot achieve a good spot landing. But lets be honest - most will cancel the landing part of RTH and manually land to avoid tip-over and be on good spot.
Nigel
Meta4
Premium Pilot
- Joined
- Jan 8, 2014
- Messages
- 15,303
- Reaction score
- 9,247
Post #19 isn't confusing anything.
It was in response to someone saying that two GPS units at the same place will show the same position accurate to millimetres.
I set up a demonstration to show what really happens.
"The trouble with quotes on the Internet is that you can never be certain that they're authentic - Abraham Lincoln"
The only way to get mm accuracy is to track phase of the downlink from the satellites. This requires that the two receivers also are phase locked to each other. In your demo with two Garmin units side by side there is no phase reference between them. Thus the only accuracy improvement is that the atmospheric phase delay is almost identical for both receivers. There are phase locked systems in use including the JPL seismic network as well as aircraft stabilization systems which put a reciever on each wingtip. These are capable of mm accuracy but are very specialized, experimental systems.
The only systems I am aware of that can measure mm displacement etc are Laser based.
My work takes me into classified areas and I can say hand on heart - GPS is not providing mm accuracy ... its just not possible.
Lets take the Geodetics guys who have just surveyed my estate and marked out the new buildings. They don't use 'Hikers GPS' ... they use sophisticated units costing an arm and a leg !
But what is used to actually determine accurate position ? They find the spot height marker nearest my land. They reference it ... then move the base machine to a point of advantage with Laser ... that gives distance moved. That base unit then using laser again is trained on the Chain boys stick (prism actually) to measure precise distance, height etc. The GPS only check-referenced initial base mark.
If a GPS could work to such accuracy - then why would a smart bomb use imagery for its closing kill ?
Star Wars aint here yet !!
Nigel
My work takes me into classified areas and I can say hand on heart - GPS is not providing mm accuracy ... its just not possible.
Lets take the Geodetics guys who have just surveyed my estate and marked out the new buildings. They don't use 'Hikers GPS' ... they use sophisticated units costing an arm and a leg !
But what is used to actually determine accurate position ? They find the spot height marker nearest my land. They reference it ... then move the base machine to a point of advantage with Laser ... that gives distance moved. That base unit then using laser again is trained on the Chain boys stick (prism actually) to measure precise distance, height etc. The GPS only check-referenced initial base mark.
If a GPS could work to such accuracy - then why would a smart bomb use imagery for its closing kill ?
Star Wars aint here yet !!
Nigel
Nigel,
Look at these references or Google NASA JPL GPS or similar.
JPL News Release - NASA Navigation Work Yields Science, Civil, Commerce Benefits
GIPSY-OASIS
By using differential GPS and keeping track of phase it is possible to get sub cm accuracy. It is not yet practical for weapons terminal guidance but probably will be eventually.
Paul
Similar threads
