Tomorrow I am going to try to make a video ( I said try ) on how I map with my EVO RTK
Mapping with the V1 RTK Unit
- Thread starter jmason702
- Start date
Detritalgeo
Well-Known Member
Awsome. I may be taking delivery of one soon... most of the places I want to map are off the grid so I guess I am PPK only. I already have the Emlid RS2 kit and an older Hemisphere GNSS if needed.
Just uploaded it now, so should be in 2.7K soon, I hope this helps some people, and feedback from enterprise guys would be great as a comparison.
This is my workflow to a degree, trying to keep it under 30 min
Looks identical except for the fact that once your drone reconnected it successfully completed the mission upload. Then it uploaded it again after you got the RTK fix back. Yours also did not say "RTK Not Ready" like mine does every time it reconnects in Mission mode. The turning behaviour was as good as I have seen but we have been having to fly in a lot of wind so it has a much harder time hitting that waypoint coordinate as precisely. Good vid man!
Good, I wanted to compare the two, and the only way I could think of was to do a video.
I haven't seen any videos on youtube about the EVO RTK in real-life mapping, so a plus there also
Dave Pitman
Well-Known Member
Dave Pitman
Well-Known Member
One thing I noticed today is that the position accuracy of the images was the same whether or not RTK was turned on or not. I flew a short mission where I had establised a FIX solution on the drone but forgot to toggle RTK ON at the top of the RTK dialog. Then, I flew the mission again, and turned it on this time.
Looking at the exif on the images and in the .mrk file, there was no difference in reported accuracy. It appears that the ON button in the dialog will just keep the craft from taking off without a FIX, but does not make any difference otherwise. Maybe one of you guys can confirm this finding.
Looking at the exif on the images and in the .mrk file, there was no difference in reported accuracy. It appears that the ON button in the dialog will just keep the craft from taking off without a FIX, but does not make any difference otherwise. Maybe one of you guys can confirm this finding.
I told one guy, that just having the RTK module on the drone you get better accuracy, seems to be true
Dave Pitman
Well-Known Member
Yes, if the RTK module is installed but no NTRIP, you still get SINGLE status precision which is much better than the regular gnss receiver on the drone.
But, if you have a FIX solution, you get the same accuracy on your images even if the the RTK button is in the OFF position.
That would be completely contrary to the software design but honestly it wouldn't surprise me either. How do you establish a fix without corrections. Mine says RTK is not ready and single solution if the RTK toggle isn't on.
Dave Pitman
Well-Known Member
I can have the toggle off and establish the NTRIP connection. They do not seem to be dependent on this firmware anyway.
What is elevation optimization? It seemed to do nothing in that mission I flew...
I would like to know what it is supposed to do, anyone try this??
I would like to know what it is supposed to do, anyone try this??
I tried it and while I see what it did, I don't understand why unless it's for some kind of internal check. Once your mission is complete, it will fly from that finishing point to the central point of the entire flight plan and shoots oblique images on the way back. My best guess is that it is creating images from a completely different perspective that may help tie all the other images together. Just a theory though. Maybe there's a functionality in some photogrammetry software that can use those images for some purpose.
@jmason702 Thanks so much for making the video, nice job, I'm impressed and appreciative.
I concur with @elphtrooper assessment of the comparison to Enterprise.
Regarding the turns of the Evo II on missions, I'm just not impressed. Compared to anything Arducopter, Evo is atrocious, and compared to Anafi on Pix4d Capture missions, Evo is bad. I haven't flown DJI, but watching others; the Evo seems comparable, maybe a little worse.
One thing I noticed in the field today is that the mission countdown timer stops during turns. Haven't tested to see if the mission estimated time is accurate or not, but again, all that time wasted in the turn sequence just eats up potential acres mapped.
Another observation is that the mission planner assumes a faster interval between photos than the vehicle seems to allow. In other words, I plan a mission at 80% overlap, 6 m/s (the max allowable speed for the experiment in question). Then I get to the field with the vehicle actually connected, and the max speed drops to 3 m/s. I happened to be at a 100ft agl limited site today, and the best compromise I could get was 72% overlap at 5 m/s. I'm wishing I could get photos every 1 second, rather than every two seconds...
I ran some numbers yesterday the Evo II RTK is about 2-3 seconds slower in each turn than the Phantom Pro with DroneDeploy. It may not seem like much but on an average site there are about 20-30 turns in DD (2 corners) so that's anywhere from 40-90 seconds of wasted time. The Evo II and Mavic 2P have a narrower field of view so there have typically been even more turns even when flying a little higher to make up for focal length.
