Increase altitude precision

Q&A forumCategory: QuestionsIncrease altitude precision
Barcaz01 asked 1 month ago
Hi at all,
I read in this forum that SimpleRTK2B with RTK (that I actually own, in my case with NTRIP) has the accuracy of 1cm + 1mm/km for horizontal and 2cm + 1mm/km for vertical position. What’s the SimpleRTK3B Pro precision with ntrip rtk for horizontal and vertical position?
I’d like to use ardusimple for ground plane leveling to help excavator operator in constitution site, so getting also a good vertical accuracy nearly 1cm + 1mm/km should be important (in this case I would setup the base nearly the construction site).
How can I achieve this? Is it possible for example to make an average of 4-5  rtk measurements? If so, how many milliseconds should pass between one measurement and another? How can I know which accuracy I get in this way? It’s to  consider that I develop apps so I could do a custom solution if needed…
Thank you for your attention 🙂
Staff replied 1 month ago

The simpleRTK3B Pro can reach 0.6mm horizonal and 1.2mm vertical accuracy, but needs to be in open sky conditions, with a calibrated antenna and with ntrip service with full constellation in L1/L2/L5. Or setting up your own base station with simpleRTK3B.Pro.

replied 1 month ago

A lot of grader / bulldozer implementations use dual-antenna configurations. One might also look at other methods of augmentation, perhaps with IMU/GYRO

You might have to experiment with your own specific use case, and proximity of the base, or second antenna config.

replied 4 weeks ago

yes but a grader / bulldozer configuration costs a lot.
I was thinking to help the person in the excavator with an operator that reads heights of the ground with the use of Ardusimple.
For sure I’ll have to do tests.
A test that I’m thinking to do is using my permanent base to send correction and help the survey-in of the base that I’ll use in the construction sites (distance up to 35km between them). Then use a rover connected to the base of the construction sites.
How many minutes do you suggest to do the survey-in?
I’m searching for repeatability of results, because surveys taken in different times of the work-in-progress have to be comparable.
Thanks for help

replied 4 weeks ago

If you are using a permanent base to supply data briefly to a more localized base it shouldn’t take more than a few minutes to get an RTK FIXED location you can subsequently push into the local base. I would recommend recording that at each cycle so confirm it is not deviating, the receiver can report an accuracy estimate, and that should be observed. Around here people generally use a semi-permanently mounted post on the worksite, or available fence post, or hard mount point, with a 5/8″-11 bolt/rod which the antenna can be spun onto.

I would not use SURVEY-IN using an unaided 3D Fix, the accuracy/repeatability is LOW.