Page 1 of 1

Longitude in NMEA output

Posted: Thu Dec 10, 2020 5:36 pm
by geonerd
I ran into what appears to be a discrepancy between the hedgehog coordinates reported by Dashboard and the coordinates the beacon outputs in the pseudo NMEA string. Hoping someone else has experience with this and can suggest a solution. I’m using super beacons with v6.24 Dashboard and drivers.
This discrepancy appeared when I set up a rectangle with tape measures measuring 5.5m in the x direction and 10m in the y direction. I placed stationary beacon 16 on a tripod precisely over the (0,0) point and stationary beacon 17 at (5.5,0). Two other beacons were close to (0,10) and (5.5,9) but these were not precisely placed. All stationary beacons were at the same height, 1.8m. I oriented the submap to have SB16 at the (0,0) point and have SB17 along the X axis at the distance determined prior to freezing the submap (I love the Dashboard feature which allows you to adjust beacons to be along the Y=0 axis; not sure if that is new, or I just noticed it, but I use it a lot now). The location of SB17 was consistent with the tape measure and SB18 and SB19 locations looked right also. The Georeferencing was set to the lat and lon here, about lat=40 and long=-105. I’m very close to the UTM zone central meridian making the convergence angle close to 0, so I didn’t bother with any corrections for that.
I used an app to record the pseudo NMEA string and later converted those to projected coordinates in m. With several beacon set ups similar to what I describe above I measured the location of the x= 1, 2, 3, 4, and 5 meter points along the Y=0 and Y=10 lines and the (0,3) point. The hedgehog coordinates on Dashboard looked about right, usually within 5 or 10 cm of the tape measure. I had to ask my wife to help taking these measurements so I could watch the dashboard screen; I’ll be washing the dishes for the next few nights.
I took the lat and lon values from the recorded NMEA strings and converted them to the local UTM coordinate system (UTM Z13N). That conversion was applied with Oasis Montaj by Seequent, reliable software I have used for this purpose to successfully convert GPS data in the past. Every time the coordinates along the X=0 axis were accurate within about 10cm, usually better. But moving east from the X=0 axis caused increasing x position errors. The y coordinates were consistent with the tape measure and dashboard. For instance, Dashboard displayed an X position of about 5m when the hedgehog was at X=5, but the X coordinate calculated from the NMEA string was about 4m from the origin. For other points the error was roughly 20% of the distance from the X=0 axis.
The system appears to be working well as displayed in Dashboard, but I wonder whether there is a problem with the longitude output in the pseudo NMEA string. The distance between latitude lines at my location is about 77% of the distance between latitude lines at the equator. That is pretty close to the error observed, but that similarity could just be a coincidence.
Has anyone else compared the dashboard reported coordinates vs the NMEA output coordinates? I’d like to know if others have experienced the same thing, or if I’m doing something wrong and somehow causing the discrepancy. Sorry for the long post, but I thought it might help to include all the details.

Re: Longitude in NMEA output

Posted: Tue Dec 15, 2020 1:03 am
by smoker77

What hardware type of beacons do you use (HW4.9, Super-Beacons or others)?
Can you share some example with wrong result:
- location in dashboard (X, Y)
- location from NMEA sentense (latitude, longitude)
- geo reference (latitude, longitude)
We will check the data and reply.

Re: Longitude in NMEA output

Posted: Tue Dec 15, 2020 5:42 am
by geonerd
I'm using super beacons.

One of the sets of measurements is below. The DashX and DashY are the distances on the tape measures. The dashboard X and Y values matched these pretty well, mostly within about 10 cm. I didn't write down the exact numbers displayed on dashboard, they were constantly changing as the beacon location was updated. The log file from dashboard may be attached to this message, but it's difficult to tell if I succeeded in attaching it. It would probably be difficult to sort out which measurements in the log file correspond to the list below.

georeference latitude, longitude
39.66633, -104.94446

DashX, DashY, Lat, Lon
1 0 39.6663296 -104.9444509
2 0 39.6663296 -104.9444411
3 0 39.6663295 -104.9444315
4 0 39.6663286 -104.9444218
5 0 39.6663293 -104.944412
0 3 39.6663563 -104.9444606
0 10 39.6664216 -104.9444586
2 10 39.6664207 -104.9444403
3 10 39.6664207 -104.9444308
4 10 39.6664205 -104.9444238
5 10 39.6664205 -104.9444143

Thanks for your help!

Re: Longitude in NMEA output

Posted: Tue Dec 15, 2020 9:08 pm
by smoker77

I just checked and it seems we have some issue in longitude calculation.
Please clarify what type of system do you use - IA or NIA?
I will send you fixed version of firmware and if it resolves the situation on your side, we will include this fix into the next release.
Thank you for your feedback

Re: Longitude in NMEA output

Posted: Tue Dec 15, 2020 10:01 pm
by geonerd
I use NIA. Thanks.

Re: Longitude in NMEA output

Posted: Wed Dec 16, 2020 5:55 pm
by smoker77
Please try update hedgehog to version from attachment

Re: Longitude in NMEA output

Posted: Fri Dec 18, 2020 3:09 am
by geonerd
The longitude and latitude in the NMEA string from the hedgehog with the updated firmware matches the locations from dashboard and my tape measures. Thanks for updating and providing that firmware so quickly!