Hei,

So, we purchased a P4RTK and the base station, and it's basically useless.

1) After connecting everything together, selecting the D-RTK2 as the RTK source, it NEVER gets to "FIX". Tried in several places, in wide open areas, nothing. The app says "weak RTK signal", forever. I can deactivate RTK and fly the missions, sure, but it very much defeats the point of the D-RTK2 station.


2) We tried the terrain awareness option, fed in a DEM tif and tfw, in ellipsoid heights, selected an area to fly, setup to flight height of 40m, start the flight....and see the drone shott way over 100m for no reason, returning images from a complitelly stupid height. Flying the flight with the "2D photogrammetry" option is fine, the drone gets to where we expect, but again, the desiered awareness is gone.

And don't get me started on the documentation...there's nothing in their to help troubleshoot issues, and only vaguely sometimes instructions on how to perform tasks.

I think the D-RTK2 is the worst of the pair, really bringing ZERO added value to the system. Which is sad, the specsheet and theorethical abilities are great....

Any news on updates of this whole mess? This is clearly mostly software issues.

Oh god. We played a bit with the input DEM for the terrain awareness, and it ended up working when we fed a DEM that had its minimum value at 0. Which meant we had to transpose it down 160m or so, to an elevation system that made no sense. We'll investigate more next week, but this is crazy! It's like somehow the drone tries to compensate for any geoid height by making sure the minimum of the DEM is 0 no matter what? Something like that? it makes ZERO sense.

Welcome to the fray. The P4RTK & D-RTK 2 definitely have their issues, and there is an abundance of threads discussing them and how users have dealt with them. You might benefit by some reading.

And I'm in full agreement that the documentation is sorely lacking.


But not everyone has reported the issues you've experienced.

It's quite infuriating that following the steps as described both by the doc and by third party tutorials fails to give results...

And the terrain follow "DEM must have 0 somewhere" just breaks my brain. Why would there be such a limitation/bias?

Right now, it looks like the purchase of the D-RTK2 was simply a mistake...

It is a lame requirement if that's the case (I have not tried terrain awareness yet). But if so, can you simply edit a single unimportant pixel to an elevation of 0, rather than shifting the whole DEM vertically by the required amount?

And thank you for sharing your experience, it helps us all.

patiam Posted at 8-30 11:09
It is a lame requirement if that's the case (I have not tried terrain awareness yet). But if so, can you simply edit a single unimportant pixel to an elevation of 0, rather than shifting the whole DEM vertically by the required amount?

And thank you for sharing your experience, it helps us all.

Hei!

Turns out it's not even that. We tried somewhere else with a source DEM that was transposed to start at 0, and one 30m upwards, and the 30m up allowed us to fly 60m above ground as planned (and 200m above the DEM). For instance, the first image has a registered Z coordinate of 251m, while the DEM reports 52m at that location. For a desired 60m above ground flight, it's not quite as expected.

We also noticed that the drone takes off and flies up to the elevation of the first waypoint, even if that flight height is VERY low. In our case, about 3m above the take off point! It's like the terrain isn't taken into account until the waypoint is reached :/  I have read somewhere else that this is a known (catastrophic) issue.

LucGirod Posted at 8-30 12:12
Hei!

Turns out it's not even that. We tried somewhere else with a source DEM that was transposed to start at 0, and one 30m upwards, and the 30m up allowed us to fly 60m above ground as planned (and 200m above the DEM). For instance, the first image has a registered Z coordinate of 251m, while the DEM reports 52m at that location. For a desired 60m above ground flight, it's not quite as expected.

Thanks for continuing to report your findings. As I mentioned I have not used terrain awareness yet but in my readings on this forum and elsewhere I have not heard of the issue you initially described (DEM needing zero or shifting for some reason). I am not doubting your experience, just noting that AFAIK you are the first to report it (which could be due to it being a rare issue specific to elevation ranges such as yours, or a firmware-specific one, or other factors). Can you please post your GS-RTK and aircraft & R/C firmware versions?

Also: to confirm, do you have RTK fixed before starting the mission in every case?

The second issue, dangerous autonomous flight at mission start when using terrain awareness has definitely been documented by multiple users. Really sounds like a poorly written algorithm there.

We just shipped the drone for some fieldwork, so I can't check the firmware. But "last" is a good bet, as I updated everything this month.

Also, no RTK. As reported above the D-RTK2 just doesn't work, so the flights are done in GNSS mode. Still, flying the same flight plans repeatedly results in very consistent paths and images, so that's not the issue. If the flight height was off by a meter or 5, o would gladly blame it on that, but when the error can be as high as 100s of meters...

If I add 10m to the DEM file, the drone flies 10m higher. So for now, it seems like I need to prepare a series of DEM in 5m increments, get to the field, try one, see how high the drone goes to, land back down, compute by how much it was wrong, select the appropriate DEM that would give the opposite bias and fly using it...

LucGirod Posted at 8-31 09:43
We just shipped the drone for some fieldwork, so I can't check the firmware. But "last" is a good bet, as I updated everything this month.

Also, no RTK. As reported above the D-RTK2 just doesn't work, so the flights are done in GNSS mode. Still, flying the same flight plans repeatedly results in very consistent paths and images, so that's not the issue. If the flight height was off by a meter or 5, o would gladly blame it on that, but when the error can be as high as 100s of meters...

Thanks for the additional info. Be aware that w/o RTK, the aircraft is relying on barometric pressure for altitude. This is generally pretty good for determining AGL  altitude relative to takeoff position, but can be far from correct wrt true HAE.

That, plus some sloppy coding in how "terrain aware" flights are executed (esp/ w/o RTK fixed), may well be the reason you require such trial-and-error finagling in order to get the behavior you desire. You mention repeatability but what is the time frame within which you are flying these repeat missions?

Just an idea.

Regarding the D-RTK 2, you may have gotten a faulty one. It's a bit quirky and far from what I would call Survey Grade gear, but if it's not broken the thing should work, if used correctly.

patiam Posted at 8-31 10:09
Thanks for the additional info. Be aware that w/o RTK, the aircraft is relying on barometric pressure for altitude. This is generally pretty good for determining AGL  altitude relative to takeoff position, but can be far from correct wrt true HAE.

That, plus some sloppy coding in how "terrain aware" flights are executed (esp/ w/o RTK fixed), may well be the reason you require such trial-and-error finagling in order to get the behavior you desire. You mention repeatability but what is the time frame within which you are flying these repeat missions?

Hmm, flying using the barometer for relative elevation would make some level of sense, but would in fact be absurd in conjunction with a DEM. It might be the reason indeed.

I guess I'll need to figure out why the D-RTK2 never gets a fix and try the terrain awareness again if I ever do.

LucGirod Posted at 8-31 10:23
Hmm, flying using the barometer for relative elevation would make some level of sense, but would in fact be absurd in conjunction with a DEM. It might be the reason indeed.

I guess I'll need to figure out why the D-RTK2 never gets a fix and try the terrain awareness again if I ever do.

Yeah, autonomous GNSS Z solution can be 10's of meters off, so non-RTK UAS generally use the barometer for altitude. It is noisy and not without problems as well, but much safer for most applications.

I know, but for typical "flat" flights, it computes the 0 for the takeoff point and works from that. Do you think it uses a fixed pressure to elevation formula for the terrain following?

I think (although I have not searched for documentation to confirm it) that DJI advises against using terrain following w/o RTK fixed, and surmise that it is b/c they are (rightfully) not confident in whatever way the aircraft is going to calculate some sort of HAE from its barometric pressure derived altitude for use against the loaded DEM.
Also, this may be of interest:
https://forum.dji.com/forum.php? ... 111&pid=2043460

Sure would be great if some actual DJI support or dev techs paid attention to these forums, eh? Especially given their p1$$-poor documentation.

Can somebody help me? About my DJI Phantom 4 RTK

My RC and RTK Mobile station does'nt connect.  
Whenever i turn on the RTK and select the D-rtk mobile station. "The SDR is in idle mode" appeared on the screen and does'nt connect at all. How can i fix the SDR idle mode, so that i can link the mobile station to the RC and start operate with RTK flight?. Please help.

You need to link the D-RTK 2 and your RC. From the User Guide:


If that doesn't help, I'm not sure what is wrong.

Perhaps the OP should edit the title of this thread...

At least half of his issues were due to operator error, not problems with the P4R. I'm a vocal critic of the shortcomings of this far from perfect system, but calling it a joke is not at all accurate. Hundreds (thousands?) of professional pilots have conducted many thousands of flights producing survey-grade results with this kit.

patiam Posted at 12-17 08:43
Perhaps the OP should edit the title of this thread...

At least half of his issues were due to operator error, not problems with the P4R. I'm a vocal critic of the shortcomings of this far from perfect system, but calling it a joke is not at all accurate. Hundreds (thousands?) of professional pilots have conducted many thousands of flights producing survey-grade results with this kit.

No, I stand by my title. The system is a huge mess with no documentation, no error messages or reports...

By doing everything by the (limited) book, my system didn't work, and would not give any meaningful or helpful feedback.

"Weak RTK signal"? What the hell does that mean? When everything is paired, a coordinate is entered for the base station, the sky is wide open and there's no other electronic nearby.

This system is a joke, as any serious system would communicate on what the issue is, it would have an in depth documentation, and we wouldn't need to scour the internet and user forums to get basic things working.

Also, it shouldn't randomly try to kill you by only starting to use terrain follow at the 1st point of the flight path instead of from takeoff.

Hi,
I agree with Patiam there are plenty of people doing precise surveys with the DRTK-2 receiver. On the other hand, I understand you, I just browsed your publications on the web, we do the same job. You will surely have more satisfaction using this: https://kartverket.no/en/on-land/posisjon/guide-to-cpos/
If your study sites are still the same you can also consider installing your own reference station. https://learn.sparkfun.com/tutor ... ference-station/all

LV_Forestry Posted at 12-22 00:27
Hi,
I agree with Patiam there are plenty of people doing precise surveys with the DRTK-2 receiver. On the other hand, I understand you, I just browsed your publications on the web, we do the same job. You will surely have more satisfaction using this: https://kartverket.no/en/on-land/posisjon/guide-to-cpos/
If your study sites are still the same you can also consider installing your own reference station. https://learn.sparkfun.com/tutorials/how-to-build-a-diy-gnss-reference-station/all

Hei,
Well, for sites in the mainland, we of course just switched to CPOS/N-RTK, and it works fine enough. And since the P4RTK exports rinex files, we can even postprocess after the facts.

The main reason we bought the D-RTK2 is that we rather often work in areas where we do not have network connectivity (remote mountain areas, Svalbard...). Then, we were hoping to set the D-RTK2 on a known pint and use it as the base, which is one of the mode that should work. But it doesn't. At least not with the hardware we got delivered. It is absolutely possible that we were delivered a faulty unit, but given the lack of documentation and feedback from the unit, it's basically impossible to know.

patiam Posted at 2021-12-15 10:01
You need to link the D-RTK 2 and your RC. From the User Guide:

Hi Captain,

We have the same problem, after took my last flight to change batteries and turn it on again, it doesn't connect to RTK mode. after i turn on the Aircraft RTK positioning and enter the RTK service type to D-RTK 2 mobile station, "SDR is in idle mode" shows up and never connect to the base station. i tried several times to update latest firmware and downgrade to lower version but it always ended up to "SDR is in idle mode" and also re-link / follow all the instruction in the manual like what you posted. but nothing happens on the linking until the battery is drowned.

Please help me because I don't how to use this P4 RTK on its purpose. i cant fly with RTK only GNSS mode makes my processing prone to erroneous value.

-How will i turn off the idle mode?

djiuser_sIyXTF434NxX Posted at 2021-12-14 23:56
Can somebody help me? About my DJI Phantom 4 RTK

My RC and RTK Mobile station does'nt connect.  

Hi mate, did you fixed your issue? on SDR is in idle mode?

What controller do you have the built in screen or SDK version?

Looks like he has the built-in screen from the pictures.

We have returned our D-RTK2 that apparently had an hardware fault, and gotten it back. We put the firmware for the Matrice and it worked very well. Now, it didn't work at all with the Phantom, I guess it's a different firmware?

But the system is still "a joke"?
Sorry you got a bad D-RTK 2 unit initially. It would be nice if that never happened, but with tech it just does sometimes. Glad you're getting better performance from the repaired unit.

Not sure but I think the concurrent P4R & Matrice compatibility may be dependent on using an older version of firmware. Newer versions will only support one or the other. I could be remembering this wrong however.

djiuser_yHjU8pBNSCnz Posted at 1-4 00:50
Hi Captain,

We have the same problem, after took my last flight to change batteries and turn it on again, it doesn't connect to RTK mode. after i turn on the Aircraft RTK positioning and enter the RTK service type to D-RTK 2 mobile station, "SDR is in idle mode" shows up and never connect to the base station. i tried several times to update latest firmware and downgrade to lower version but it always ended up to "SDR is in idle mode" and also re-link / follow all the instruction in the manual like what you posted. but nothing happens on the linking until the battery is drowned.

Just got the same problem with my P4 RTK. nothing solves the problem, downgrade, upgrade, different RTK stations, changed between my 3 P4 RTK but the remote controler is the problem. 2 controlers works fine with 3 aircrafts, but one controller does not function with any of these 3 aircrafts. I`m pretty sure the problem is with the controller, but don`t know where?

Maybe you can try our RTK Base Station

RTK Base Station

penny1996 Posted at 6-13 17:59
RTK Base Station

https://www.sveav.com/

penny1996 Posted at 6-13 17:59
RTK Base Station

https://www.sveav.com/