Advanced Question Inertial vs Surface Reference for Navigation

[rcraig42]

I've been experimenting with suborbital "atmosphere skipping" ballistic flights in the XR-2, starting the "glide" at around 55-75 km altitude, and Mach 15-20.

One thing that I quickly found out was that while the Map Mfd and HUD give the surface based great circle bearing to the targeted base, the bearing that actually works for these flights is the inertial bearing, taking into account the Earth's rotation beneath your flight path. That is, for a 1 hour flight you need to move the target about 15 degrees to the east (how much the Earth rotates in that hour) and calculate the great circle bearing for that.

My questions are, how does the transition from surface to inertial based navigation work in Orbiter (the air does move along with the surface in Orbiter 2010... right?)?

btw, I realize it is always inertial outside any atmosphere, should one exist. But I don't know at what range of dynamic pressure you transition to a state that is no longer riding along with the atmosphere. Or even if orbiter takes riding along with the atmosphere into account at all. I don't have the patience to try a long distance, lower altitude flight in one of the airplane addons to attempt to find out.

and

How does that compare with reality?

 

[RonDVouz]

I just think the air density changes at altitude until the point where aerodynamic control is useless but to scale isn't an accurate depiction of the real air densities at altitude. Orbiter doesn't take into account air pressures or wind. The simulation operates on the idea that the weather is perfect every day and barometers don't exist.

 

[ADSWNJ]

I too would love to know the answer to this. I could rationalize that the air mass close to surface level is travelling approximately the same speed as the ground (wind ignored), and thus the ship tracks to the surface. (Eg. Fly at 50m at 100m/s due north from S30 E0 to N30 and you should still be at E0). Flying from the same start due North at successive altitudes up to 200km (adjusting thrust to maintain alt), and I suspect you will gradually slip off the surface meridian. How much though is a great question.

(Alternative model ... if tracking a 000 azimuth locks the ship to the meridian, then how much East heading do you need to compensate for Earth rotation).

My interest is for Glideslope 3 (future version) and figuring out a whole new HAC solution based on turn authority at different speed, AoA and pressure density, versus the final alignment. And academic interest too!

(Another version ... if you fly across America on a real plane, then the East direction is always faster usually due to Jetstream winds. But imagine winds were zero at all altitudes. Would the East journey be longer, shorter or the same?)

 

[rcraig42]

For the flights I was talking about, the West vs East is an interesting question. It still seems to consider airspeed (which should = GS?, never checked that) for drag purposes, but the distance is less going West considering that the Navigation has the target moving toward you. So it should take less energy or time to go West.

Unless of course it "docks" you for your starting orbital speed according to altitude/velocity I need to check that out when I get my power computer back working.

It seems to me like an atmospheric flight with no wind, even at 55-75 km, should be the same both ways.

---------- Post added at 09:45 AM ---------- Previous post was at 09:31 AM ----------

btw, I downloaded the A380 to attempt a long flight at 10 km altitude to see how it tracks. But there is no wing leveler to keep its track through the air stable, and it isn't stable on its own. so no way to tell if it tracks with the air/ground or not.

---------- Post added at 09:49 AM ---------- Previous post was at 09:45 AM ----------

one more thing, the author of Aerobrake MFD must know the answer to this, the projected ground track on that is flawless.

 

[ADSWNJ]

one more thing, the author of Aerobrake MFD must know the answer to this, the projected ground track on that is flawless.

Well - Aerobrake comes with the source code if you like a challenge. Jarmo Nikkanen and Gregario Piccoli are the authors (per ABMFD.cpp). The code looks pretty clean, with descriptive variable names, etc, but the comments are very sparse. It would be a labor of love to pull it apart to determine the logic. There's a Runge Kutta method right in the bowels though .