Support IGS SP3 files about rinex

[gwbres]

Hello @ljbade,

SP3 parser should be a dedicated crate since it's not a RINEX and it differs a lot.
Eventually when it becomes available, we should add an --sp3 argument to the command line tool to make it possible to provide such information.
Similarly to the --nav argument, to combine two RINEX for example

[gwbres]

Hello @ljbade,

I iniated this library, that we will merge into the georust community once stabilized.

It will be easy for the rinex-cli tool, through the rinex::quality::QcContext , to take one or several SP3 into account.
My understanding is, this helps solve PPP. The rust RINEX ecosystem is quite far from that at the moment, so my understanding is, this will not impact before a very long time.

They're might be an interesting aspect if we support real time parsing of all of these IGS datasets, mainly for real time PPP solving, which needs SP3 predictions as soon as they are published.

Feel free to contribute to all of that, lots of topic and I can only achieve so little by myself, especially considering this is no longer part of my main occupation.

[ljbade]

@gwbres nice, I will take a look.

SP3 is useful when you want precise orbits, so is often used in PPP but can be used for a number of other things. There are even some SP3 files which just contain the "broadcast" orbits translated from the RINEX NAV files, which I found a useful reference when writing my own satellite position calculator.

[gwbres]

Hello @ljbade,

SP3 is definitely interesting data to have. We'll soon have the ability to load an SP3 pool into the QcContext just like in teqc.

The main thing it needs now is an interpolatin method, just like here.
So RINEX pool can compare to the provided SP3 context, and also to combine SP3 together and form a seamless SP3 context.
rust::nyx-space::Interpolators might prove handy.

[gwbres]

@ljbade,

The current PR already has the ability to load a pool of SP3 files.
This allows loading a context spanning several days at once.

We currently only plot SP3 orbits and clock data, there is no processing involved.
I will first conclude SP3 interpolation within the SP3 crate. Once that is done, it will be possible to correlate SP3 data/predictions, to RINEX NAV.

They say Clock predictions should not be interpolated, although they seem to be very linear and stable.
Are you familiar with Clock predictions processing ?
So far my main reference has been ESA NaviPedia