relative accuracy cogo tools

[Duane Frymire]

Wondering how to use this tool, I thought maybe it would pick up bad solutions for me. I had an rtk position I knew was bad, so put it in list with good ones and still got 0.06 positional tolerance. Went back and shot the point with total station off two good rtk position nails and confirmed point was 1.4 feet off. So, not sure the 0.06 is a true representation of the points I know are pretty good; they might not be that good. I looked in manual and don't find anything.

On related note, are there instructions or tutorial on how to export g files for use in least squares analysis (Carlson survnet in my case). Can this even be done with rtk or does it require base or cors processed points? Sorry, not sure what the g file is/represents even.

 

[Jim Frame]

Can this even be done with rtk or does it require base or cors processed points?

Yes, it can be done with RTK, both network and local base. I'm not familiar with Carlson SurvNet, so I can't offer guidance on that. But the g-file export works great with Star*Net.

 

[Duane Frymire]

Thanks Jim.

 

[Matt Johnson]

Relative accuracy is not a tool to detect bad RTK initiations and fixes. It only provides statistical error estimates between two points based on the covariance matrixes stored for each point. Bad RTK initiations are not reflected in the covariance matrixes. You need to use the Boundary profile to verify good RTK fixes.

 

[Duane Frymire]

Thanks Matt. Yes, I already knew it was not a good position. 25 minutes and no verified rtk solution, and bcp was over a foot from stored rtk (non-verified). So this was essentially a 25 minute rapid static observation (I saved the bcp solution). Just exploring usefulness of the tool. Although the max spread of the covariance matrixes would be considered in the report output I guess, it's not really what I'm looking for.