GNSS – The Rest of the Picture
Matching the GNSS solution to the project continues to be an issue. In the previous article we looked at projects that were widely varying in requirements. First we reviewed the most general case requiring positioning to only a meter or more. Next we saw how an RTK system may be an appropriate solution for UAVs with the LOS restriction and generally clear sky. Lastly we examined the survey grade problem with only a few centimeters allowed. There are still more situations to address. What about all those projects that vary from 10 cm to 50 cm positioning? What about those that require sub-centimeter positioning?
Let’s start by looking at the options for 10 to 50 cm positioning. This is the area that can be kind of tricky because several systems can deliver approximately the same accuracy while varying drastically in cost, implementation, or range.
The most readily available system for near global coverage is to purchase an SBAS solution such as Trimble OmniSTAR or NovAtel CORRECT. These claim to be able to achieve 4 or 5 cm accuracy. However, be careful reading the specifications – that’s probably at 1 sigma (or 68% of the time). Still, these are remarkable results and often times perfect for many applications. They allow for the freedom to cover huge corridors without the hassle of coordinating three or more base stations throughout the collection period. Start-up time, how each system degrades during outages, and recovery after GNSS outages will vary with each system. These factors all need to be fully understood along with your project requirements to know whether these systems will work for you or not.
Another means of decimeter level positioning is to use local networks. These are obviously only useful in areas with established and reliable working networks. The U.S. is largely covered in this respect by one means or another (private or government). These systems can utilize even L1-only systems (meaning much less expensive receivers) to achieve these results. Again, it is wise to inquire or experiment in regards to how these systems perform in GNSS outages.
One last thing regarding the published or advertised specifications. Be sure to carefully consider the horizontal versus vertical positioning. There is a difference. All too often we read positioning and think it is a 3D reference when in fact the specifications are quiet in this regard. Usually the reference is to horizontal positioning only because the vertical positioning accuracy will never be as good as the horizontal positioning, and everybody only wants to advertise the best. Fortunately, once you know this you understand that almost all marketing literature makes this same implication. The same holds for the 1, 2, or 3 sigma (68%, etc.) confidence interval, usually it’s 1 sigma unless otherwise stated.
So if you require decimeter positioning over huge areas you have choices, but they are not easy. It is not much unlike global travel with a cell phone. You can pay local charges, by the local service ahead of time, use Wi-Fi, or buy a sat phone. If you are in the middle of nowhere with no infrastructure, a sat phone will likely be your only choice. Choosing what works depends entirely upon the situation you find yourself in.
The last area to concern ourselves with is sub-centimeter positioning. Some claim results better than 5mm in horizontal and vertical positioning. While this may be possible, and people may have jobs where they have obtained such results, I would not count on getting results like this in the real world. These results are at the very edge of technology. If you need results at this level of accuracy, it is still most likely wise to stay with the tried, true, and accepted means of surveying with optical/mechanical devices such as total stations and levels. This is my opinion and it is subject to change – but not without a lot of substantiating evidence.
This ends the discourse on GNSS options available. Until next time, check out our new UAV LiDAR video at www.lidarusa.com.