I don??t think someone should be overseeing the use of GNSS without knowing how it works.?ÿ
Could you define that? Do you mean design the receiver and antennae, be able to write your own code to calculate your position, create your own ellipsoid model, create your own geoid model, etc etc?
Because if you can't do that at some point it is a black box and you have to trust the manufacture. I would posit the same was true of the robot you are using. I am not trying to refute your statement, but asking for some clarification, some additional information.
We can all agree that everyone should "know enough of how it works..."
It would be interesting to know what people consider to be, "enough."
For 99% of surveyors, there's no need to know how to write processing and filtering algorithms to compute baselines, but I would argue that anyone using GNSS for professional-level work needs to at least know what errors must be considered and how they contribute. Time errors, atmospheric errors, orbital errors, integer ambiguity, noise, multipath.
Some of these can be mitigated by proper field procedure, others by processing methods. I'd say that management of those errors, and consequently professional-level work, is contingent upon knowing those error sources and what to do about them in order to obtain a desired result. From setting the mask angle in the receiver during collection, to knowing how to get a true independent measurement, to knowing what the difference is between the rapid and precise ephemerides when post-processing, so on and so forth.
That's the operation part of GNSS. The other critical piece is the reference system part, and that's comprised of geodesy, coordinate systems, map projections, and knowing historical datums and how they relate to each other.
Write your own geoid? Nah, not necessary. Know how geoids work with ellipsoids, what they are representing when applied (as in difference between orthometric and tidal datums, or "MSL"), deflection of the vertical, and which geoids to use with which reference frame? Absolutely.
Develop custom coordinate systems? Not absolutely necessary. But we need to know how projected coordinate systems work, how they are defined, their benefits and drawbacks, how to mitigate distortion, etc. (Although once you learn that you will be well-equipped to design your own CS.)
Analysis of GNSS vectors, positions and adjustment of GNSS data also requires knowledge of statistics. In fact, I'd argue that since both GNSS and least squares analysis are ubiquitous these days (not to mention that total station observations require statistical analysis as well when combined with GNSS), statistics is critical for anyone wanting to operate at the professional level with GNSS.
Anyone think that's too much to ask? This is in addition to the boundary knowledge that is also critical. One doesn't preclude the other.
A doctor doesn't have to design their own tools to be a professional. But they damn well need to know how the tools operate and how to get the best results from them. We should too.
