I have searched quite a bit and have not found an answer yet to this, so I apologize if it’s already out there.
I’m considering a DJI M300 drone with the P1 45mp camera payload. I really dislike the DJI RTK unit and I’d much rather have PPK anyways since I work in pretty difficult areas (mountain canyons, etc.). I’ve found instructions for linking the M300 and RS2 for NTrip but I often won’t have internet on project sites and again, would prefer PPK.
If I buy an RS2, is it possible to use that as a base to collect PPK and integrate that with the M300/P1? Do I need any additional hardware on the drone? If this is possible, are there any instructions/articles?
Yes, the RS2 will collect raw data that can be used for PPK with the M300/P1.
No additional hardware is required but you will benefit from software that will apply the leaver arm/IMU corrections to the data - Suggest looking at Klau or redcatch toolbox for this.
As far as I know, DJI has a PPK Cloud that allows geotagging images in PPK. We haven’t tested it ourselves, so I can hardly say how smoothly you can get the data. From the Reach RS2 base side, you can record raw data logs in RINEX format.
If your drone’s camera has a hot shoe, you can integrate it with Reach M2. It will allow you to calculate precise image centers without establishing a real-time connection between the base and rover. To learn the detailed workflow, you can explore our PPK mapping with Reach guide.
Update for everyone as I was finally able to test this in the field today. On the DJI M300 you don’t need to do anything special; it will automatically record a .mrk file and a .bin file in addition to the not-so-accurate geotags in the photos themselves. As for the RS2, simply set it up before the flight (I do it as the first thing upon arriving on site), set it to log the Rinex (you can do most formats although I did .ubx). Keep it logging until after you’re done with your flights, then transfer the log file so you have it.
Back at the office use a software like @DanM suggested - Klau PPK or REDcatch toolbox. I tried REDcatch as they were super responsive and cheaper. Klau was also pretty responsive but significantly more expensive. Anyways, at the office you simply start REDcatch, select the Drone profile and the output, then import the images, import the .mrk and .bin from the DJI files, and import the Rinex .ubx from the RS2. The program does the rest for you…you can have it copy the images and edit the exif geotag data, or you can choose different outputs. For example, I use the Pix4D option - the software calculates corrected coordinates for all of the images and creates a new Pix4D Mapper file with the images and inserts the corrected geolocation info. So then all you need to do is open the Pix4D file and start processing!
FYI, this page you are looking at is referring to to the D-RTK DJI base station I think…that base station can only record RTK or PPK if I recall correctly (and I think the PPK may be more of a rumor/beta than a true feature option…not entirely sure). If you use the workflow in my “Update” long response above, you don’t need the DJI RTK base station, just the drone itself and the Emlid Reach RS2 (and software like REDcatch). Hope that helps.
In this case, you are assuming that your base is set over a known point. Or at least you have solved for your Base before inputting into REDCatch? Does REDCatch ask for the Base coordinates?
That’s good news. I also have the M300 P1 RTK2 base combo and looking for solutions. I assume you did have the RTK2 connected to the M300 while collecting data from the RS2?
No, no DJI RTK base is needed. Set up the Emlid RS2 before starting the drone operation. If you have a known point, set it up on top of that. Start the RS2 recording logs before drone takeoff, and make sure it records during the flights and don’t stop it until after your last flight is completed. The M300 drone records it’s own logs and using a software like RedCatch, you can PPK (post process) the drone’s photo exif data. In RedCatch, you have to import the photos, the RS2 logs and the M300 logs.
I’d like to add a few words. You can use both RTK2 and Reach RS2 as a base for the DJI M300. A good note is that you should record the log on the base during the whole flight of the drone and that you need to place the base on a known point.
Hi Elena, Trying out the Emlid Studio beta version and the number of time stamps doesn’t match the number of photos. What would cause that? I’m using beta 10.
I guess I should mention I’m using CORS file for the base station not the RS2. Just trying to work through some flows right now.
Thanks
What is the current receiver firmware version, and what Emlid Studio version do you use? We have encountered a similar problem before. However, we released Reach Firmware 27.1, which includes the fix of the issue with missing time marks in August. We also fixed this in Emlid Studio.
Hi Elena,
I’m running v 27.1 on the RS2 and not sure what version of the studio just say Beta 10. OK maybe why this isn’t working for me. My rover file is coming from the M300 and the base is coming from the CORS network not from the RS2. I’m flying another site today with the RS2 set as my base so may give it another shot with that.
Thanks for the response.
Thanks for your clarification! It all makes sense now. The thing is that Emlid Studio can currently post-process only time marks recorded directly into the raw data log. However, DJI drones record them in a separate file, so our software just doesn’t find time marks in a log.
We understand that we have many users who can work with DJI drones, so I’ll be sure to note this as a future request.
I have some questions Elena. I’m using the the M300 with the RTK2 and the RS+ to record my GCPs. I have used the RS+ with NTRIP to record a known point. I use this point and input the coordinates in the advanced RTK settings of the Pilot app. When finished mapping I have 4 files on the SD card in the P1 camera. They are Timestamp.MRK, PPKRAW.bin, PPKOBS.obs, and PPKNAV.nav. I’m having trouble trying to use the workflow in REDcatch. I’m working with them now and they seem very helpful. We are sending them a dataset with all my RS+ GCP information and they are going to look at it. Do I need to be logging anything in the RS+ while I’m flying. I guess I’m missing something.
I haven’t had a chance to work with REDcatch software. So, I’m not an expert here. But in general, if you use Reach RS+ for GCPs, you don’t need anything but CSV with points coordinates.
Yes and NO. I can use the RS+ to get cm and sometimes sub cm readings. I have no issue exporting the CSV file and importing them into the Agisoft software. The only thing that I am having issues with is that the GCP locations are sometimes 10 to 15 meters above ground or at least above the map. The lat and long positions are 99%+ accurate. After doing a test several days ago I did get the GCPs to show about 99% accurate elevations although I’m not sure what I did to achieve this. I was using the same system and settings as a map several days before but with a completely different outcome. I know that the RTK2 records coordinates in WSG84. My NTRIP service provider uses NAD83(2011). There are several vertical datum options that I can use and have tried all of them. I guess there is some setting in Agisoft that I haven’t found yet but I have tried several relating to the coordinate systems including the one in Agisoft where you can use the coordinate system that you used for the GCPs. I think my biggest question is how the two coordinate systems interact with each other and IF they are not syncing with each other then which one do I believe? Is the map too low or are the GCPs too high? The software you can manually adjust the GCP coordinates but that doesn’t seem right. ALSO, I sent a small dataset to REdcatch and I was asked to also send the base station data. All I have is the files that the P1 records and the CVS file from the RS+ for the GCPs. Not sure what he wants. Thanks for your reply.
The guys at REDcatch advised that I don’t need their software workflow using the RTK2 BUT they suggested using the RS+ to get better more accurate coordinates and insert those into the RTK2 advanced settings. I just have to figure out why my GCPs are off. I’ll continue to work on it.
