General Question How do real world space vessels orient themselves? (How do they "find" prograde?)

[blixel]

How do real world space vessels orient themselves? (How do they "find" prograde?)

Someone asked me this question a week or so ago in a comment on one of my videos, and I couldn't answer their question. This is something I've wondered myself. I did a couple searches in the forums and didn't find anything. (However, it's the kind of question that may have been asked using different keywords.)

When the Space Shuttle (or some satellite) orbits the earth, what instrumentation do they use to orient the vessel for the various burns? How does the vessel know where retrograde is, exactly? Gyroscopes immediately come to mind, but I can't really think of how a spinning device would actually help the vessel know where the retrograde position is at.

Also, is there any difference between knowing where prograde/retrograde is at while orbiting the earth, and knowing where those key orientations are when you are on the way to the moon, or on the way to, say, Mars?

 

[Pipcard]

I think they have some sort of star tracker.

Actually, look [ame="http://en.wikipedia.org/wiki/Attitude_control"]here[/ame].

I'm not exactly sure how prograde or retrograde is found, though.

 

[Urwumpe]

Primary tool is inertial navigation = gyroscopes and accelerometers. This is today usually augmented by GPS. The computers on-board calculate a state vector of the vehicle, position and velocity, that is usually updated with more precise information from ground station measurements (Radar).

Apollo used Sextant and telescope for navigation, together with computers on the ground and radar measurements.

But for finding prograde, there are also other solutions tried: Soyuz/Progress for example use ion flow sensors, for finding pro- and retrograde.

Backup is always using your own eyes. You look down at Earth and look where you are going.

Also: It is not just one "spinning thing" in a Space Shuttle IMU, but actually 3 gyroscopes and 3 PIPA acceleration sensors, resulting in 6 degrees of freedom. These sensors are installed on a stable platform, that is kept in the reference direction by four gimbal rings. The angles of these gimbals tell you the spacecrafts attitude, the PIPAs measure the velocity changes since the last time you read their data. Three rings would be enough, but then you can get the problem called "Gimbal Lock", when two rotation axes of the gimbals are parallel and can't be told apart anymore. A fourth Gimbal is added to prevent this by always preventing the gimbal axes from becoming parallel. The spacecraft rotates, but the stable platform is always fixed in space.

A cheaper modern version is the strap-down IMU: It uses no longer mechanic gimbals and stable platforms, but instead mathematics. They are much cheaper, but less accurate.

 

[BruceJohnJennerLawso]

Primary tool is inertial navigation = gyroscopes and accelerometers. This is today usually augmented by GPS. The computers on-board calculate a state vector of the vehicle, position and velocity, that is usually updated with more precise information from ground station measurements (Radar).

Apollo used Sextant and telescope for navigation, together with computers on the ground and radar measurements.

But for finding prograde, there are also other solutions tried: Soyuz/Progress for example use ion flow sensors, for finding pro- and retrograde.

Backup is always using your own eyes. You look down at Earth and look where you are going.

And in a more general sense, everything is deconstructed against a "REFSMMAT", a sorta-inertial reference frame aligned with background stars. The Apollo missions more specifically used doppler analysis of the radio transmissions to pinpoint position & velocity to within decent accuracies.

 

[C3PO]

But for finding prograde, there are also other solutions tried: Soyuz/Progress for example use ion flow sensors, for finding pro- and retrograde.

The Luna E3 used ion flow sensors.
And the Soyuz still has the optical 'periscope' if navigation goes completely down. It would be good enough to make a manual reentry burn.

 

[Shifty]

Seems like all you'd need would be radar ranging to or from one station on the ground. Measure range and direction at two times, correct for the Earth's rotation, and draw a line to give you direction and speed.

 

[4throck]

Visual worked for me on Project Mercury. Used to do those missions on manual, VC view only and no fancy MFD.

You simple see the Earth passing below and track the features.
Straight line movement » prograde / retrograde.

Once you establish that, you can use the Earth's horizon to get your attitude right. Same as in an aircraft's artificial horizon .

 

[Urwumpe]

Seems like all you'd need would be radar ranging to or from one station on the ground. Measure range and direction at two times, correct for the Earth's rotation, and draw a line to give you direction and speed.

Good idea... if there wouldn't be a tiny detail: The radar data has big errors in it. You usually need at least three measurements to get at least a coarse state vector.

And no, just drawing a line does not work. If you measure it once on this side of Earth and once on the other side of Earth... did it use the transearth tunnel?

 

[N_Molson]

Once you establish that, you can use the Earth's horizon to get your attitude right. Same as in an aircraft's artificial horizon .

It's basically what they were using during the first Soyuz missions. The VZOR periscope had a set of smartly designed sights that allowed not-so-rough manual orientation. That wasn't accurate enough during the night side pass, though, because it is very hard to find visual clues. Especially in the 60's, when you had much less nightly city lights. So yes they had that Ion-sensor things, that overally proved unreliable. Also the ionic stream proved more complex than expected in LEO, and was affected by various magnetic events like solar flares...

Nowadays I guess it is even possible to use satellites like GPS or TDRS and an onboard computer to have an almost-realtime update of your position & orientation (if you have well-placed directionnal antennas, it is in theory possible to get the orientation of the spacecraft. Which could be useful to re-calibrate the INS from time to time.

 

[blixel]

Here's a quick scenario you can play with to test your ability to find Prograde without using any instruments or MFD's.

EDIT: I added one more ship to the scenario for a Earth to Moon flight. (I had to change the MJD for this to work.) Finding prograde relative to the sun while drifting between the earth and the moon is a lot more challenging.

BEGIN_DESC
END_DESC

BEGIN_ENVIRONMENT
  System Sol
  Date MJD 55725.3846880108
END_ENVIRONMENT

BEGIN_FOCUS
  Ship GL-Earth
END_FOCUS

BEGIN_CAMERA
  TARGET GL-Earth
  MODE Cockpit
  FOV 50.00
END_CAMERA

BEGIN_SHIPS
GL-Mercury:DeltaGlider
  STATUS Orbiting Mercury
  RPOS 2536357.94 -165855.65 -713567.96
  RVEL 784.114 21.083 2780.280
  AROT 175.08 53.46 149.47
  VROT -53.59 -20.46 -16.75
  AFCMODE 7
  PRPLEVEL 0:0.554057 1:0.989410
  NAVFREQ 166 484 0 0
  XPDR 0
  TRIM 0.033887
  AAP 0:0 0:0 0:0
END
GL-Venus:DeltaGlider
  STATUS Orbiting Venus
  RPOS -5154182.51 -0.52 3538631.49
  RVEL -4080.081 -0.000 -5942.796
  AROT -179.14 57.29 130.65
  VROT -21.86 2.95 85.70
  AFCMODE 7
  PRPLEVEL 0:0.554057 1:0.985425
  NAVFREQ 166 484 0 0
  XPDR 0
  TRIM 0.033887
  AAP 0:0 0:0 0:0
END
GL-Earth:DeltaGlider
  STATUS Orbiting Earth
  RPOS -4563554.49 -368309.61 -4716728.06
  RVEL 5559.127 569.776 -5423.091
  AROT -15.90 35.32 81.93
  VROT -40.69 36.77 27.40
  PRPLEVEL 0:0.554057 1:0.989914
  NAVFREQ 166 484 0 0
  XPDR 0
  TRIM 0.033887
  AAP 0:0 0:0 0:0
END
GL-Moon:DeltaGlider
  STATUS Orbiting Moon
  RPOS 289907.51 -1520500.85 -824489.17
  RVEL 1609.810 409.951 -189.871
  AROT 2.58 32.22 -45.49
  VROT -50.76 -25.05 -79.83
  PRPLEVEL 0:0.411546 1:0.981479
  NAVFREQ 166 484 0 0
  XPDR 0
  TRIM 0.037162
  AAP 0:0 0:0 0:0
END
GL-Mars:DeltaGlider
  STATUS Orbiting Mars
  RPOS -632529.78 -1888.61 3533836.44
  RVEL -3399.854 23.833 -608.536
  AROT 119.26 -10.76 92.07
  VROT -25.60 -28.41 104.20
  AFCMODE 7
  PRPLEVEL 0:0.554057 1:0.988305
  NAVFREQ 124 484 0 0
  XPDR 0
  TRIM 0.033887
  AAP 0:0 0:0 0:0
END
GL-EarthToMoon:DeltaGlider
  STATUS Orbiting Sun
  RPOS -21552917517.91 -2583277.81 -150645372081.02
  RVEL 28815.161 -6.109 -5391.605
  AROT -95.72 -56.29 32.46
  VROT 10.87 -13.16 31.44
  PRPLEVEL 0:0.443832 1:0.991378
  NAVFREQ 166 484 0 0
  XPDR 0
  TRIM 0.033887
  AAP 0:0 0:0 0:0
END
END_SHIPS

BEGIN_ExtMFD
END

I don't know how to do any cool LUA programming like dgatsoulis, but here's how you can use this scenario.

When the scenario loads, the HUD will be OFF, there won't be any MFD's available, and the Deltaglider is spinning out of control. While the HUD is off, orient the vessel to where you THINK prograde is at, and then press CTRL+H to turn on the HUD to see how close you are.

You may press KILL ROTATE to stop the spinning, but obviously you can not use the other autopilots.

Once you've pressed CTRL+H to turn on the HUD to check how you did, you can then press CTRL+H to turn it off again, then press F3 to select a different Deltaglider (orbiting Mars, Mercury, Venus, and the moon) and try again.

Do not press F8 to bring up the 2D panel either.

On my second attempt, I got pretty close:

My pitch was a bit closer with Mars, but I was still facing inward quite a bit.

I was further off with Venus, presumably because it doesn't have any craters or other features to watch pass by.

EDIT: In this screenshot, I'm trying to find prograde relative to the sun while going between the earth and the moon. I looked around until I could find some of the other planets. I used those planets to help me figure out where the ecliptic plane was (roughly). Then I rotated until was facing straight at the sun, then I rotated until it felt like I was 90 degrees away from the sun.

 

[Keatah]

Would an ion flow sensor be the same thing as this?
[ame="http://en.wikipedia.org/wiki/Magnetic_flow_meter"]Magnetic flow meter - Wikipedia, the free encyclopedia[/ame]