Latest technology at affordable price
by Geoff Peck, Ugly Gully Orienteers
I’ve been making foot-o maps for over 50 years now, and I think my latest GNSS acquisition is the biggest “game changer” I’ve seen…. sure, LIDAR is also a huge game changer, but there are still areas that don’t have LIDAR coverage whereas there is GNSS satellite coverage almost everywhere!
My GNSS history
I’ve been using GNSS devices since they became small enough to carry on foot, when the USA launched their GPS system. They have really helped in the production of orienteering maps, but still limited by their inaccuracy down under. I’ve found that most devices, including “data loggers” with better views of the satellites, are really only accurate to +/- 5m at best, and a lot less under tree cover.
They have been very good for general position and plotting isolated point features in the middle of nowhere, but can’t be used alone for “relative position” work where compass and distance methods still have to be used.
What has changed is that more satellite “constellations” have been launched (GLONASS, Beidou…). More recently the European Galileo system is now fully operational. Apart from increasing the number of satellites to obtain a fix, Galileo uses far better technology than the decades old GPS system, so is far more accurate.
In addition, there is now a network of reference “ground stations” (RTK) across Australia that can provide corrections for errors, such as cause by solar flares etc, in the raw satellite data. These corrections can be downloaded over the internet (NTRIP) and applied to the satellite calculations to increase accuracy.
Until recently the equipment required to use this new technology has been expensive (and bulky). However, a Spanish company has recently manufactured a small Arduino “board” featuring the newest U- Blox F9 multi constellation GNSS chip with RTK/NTRIP capability and a Bluetooth module. It was relatively easy to set up, although I am grateful to Peter Mousley for his frequent expert assistance and for explaining how RTK/NTRIP worked.
My GNSS present
I have been using this device for almost a year now and am truly impressed with its accuracy. When using NTRIP for corrections, the accuracy is in the order of 2-3 cm … yes, not a typo, that’s centimetres!
But this requires an internet connection which is not always available in remote bush (and sometimes not even available in urban areas). Also it is only really usable within a reasonable distance of a ground station. But where available, having RTK/NTRIP has made it easy to measure the “native” accuracy of the GNSS.
Without RTK/NTRIP, the native GNSS has proved to be accurate to +/_ 1m in all situations, even under tree cover. Anyone who has been involved in mapping will know that 1m accuracy is more than good enough for our sport, and so it has proved.
Proving my GNSS device
My most recent project has been the Broadwater State Forest granite area. There was no RTK/NTRIP available, but I was lucky enough to have fairly good LiDAR coverage which I was able to use to check GNSS accuracy. There was 100% agreement between GNSS position and LiDAR point features at all times. So I was able to accurately plot all the boulders using the GNSS position.
More than once I marked a boulder as ‘medium boulder, S side’ then without realising returned to the same boulder from a different direction and marked it as ‘medium boulder N side’. The difference between the two positions was almost exactly the width of the boulder. When I walk back and forth along a vehicle track, keeping to the left, the GNSS shows two parallel tracks.
This GNSS gear
Apart from the Arduino board, there needs to be a dedicated remote aerial and Bluetooth module (which come as a package from the supplier), a metal “ground plane” to reduce multi-path effects, and a power supply (I used a standard USB external charger).
The Bluetooth module sends position data to a smart phone, where I use the Bluetooth GNSS app and one of Peter Effeney’s MyOmaps apps to record tracks, waypoints and display a KMZ version of the base map. Thus I can see exactly where I am when surveying.
The accuracy of all GNSS systems depends on the signal strength from the satellites. Avoiding your body getting in between the GNSS device and satellite ‘line of sight’ is one obstruction you can easily overcome. So mounting the antenna on top of your hat is essential. Unfortunately that almost guarantees getting some multi-path interference (reflected signals from the satellites). A ground plane keeps the interference to a minimum. I use the top of an old metal tin, about 14cm diameter.
Acquiring the gear
I purchased the main Arduino board (simple RTK2B) from ArduSimple. I added their multi band GNSS antenna – ANN-MB-00 (Galileo works on a different frequency to other systems). All for around 200 euros (A$320 at time of post).
To this I added a Keyestudio XBee 2.0 Bluetooth module (which just plugs in to the board) and made myself a plastic box to keep it fairly safe.
- Ardusimple simple RTK2B board
- Multiband antenna
- Bluetooth module
- Ground plane (metal plate)
- Hat (not necessarily from Bunnings, but works well!)
- Power supply (or battery pack
- Bluetooth GNSS app from clearEvo.com on playstore
- Tracking app such as MyOmaps, Maprun etc
From field to computer
Regrettably I’m an ‘old school’ mapper so prefer to use this technology to help me draw my survey details on drafting film in the field. I then transfer that detail, plus GNSS data, into OCAD when I get home, You may prefer a set up with tablet to draw directly into mapping software.
If you do go down this track, then please do let us know how you get on. Geoff will gladly field queries at joffpeck at hotmail.com. Feedback and reports are welcomed by Ken via this website or by Geoff.