Air density as a function of elevation

[bqpd]

This would lead to more-meaningful sensitivities for aircraft.

[bqpd]

Concave (i.e. works if rho and h are good) to 27.4km:

(model from https://en.wikipedia.org/wiki/Density_of_air#Altitude)

[pgkirsch]

wait i did this already!

[bqpd]

woot!

[bqpd]

And here I was looking forward to using gpfit

[pgkirsch]

https://github.mit.edu/convex/GPjet/blob/master/standard_atmosphere.py

[pgkirsch]

I tried using GPfit but it failed me for some reason and I haven't had time to figure out why

[bqpd]

I'll mess around a bit some other time, then :D

It doesn't look like it'll need Signomials...

[whoburg]

wait that only goes to 15km, a bit higher would help my current application

also, a GP, not SP, model seems achievable from the plot above, assuming that the application is putting upward pressure on h and rho

[bqpd]

In such a model I'd be tempted to bake gravity in as a EQ constraint just so it'd stop showing up in sensitivities :p

[pgkirsch]

Hmmm that was just my first take at a working solution, will look into it some more.

[whoburg]

great! awesome.

[pgkirsch]

Addressed in #45