Spacecraft orientation and navigation

[Turbinator]

Now here is a interesting question.

How does a spacecraft know if it is pointing prograde, or retrograde, or in any other orientation? I know the star tracker is used to determine where the spacecraft is relative to the universe, and it's relative orientation. However, how does it know it's pointing in, for example the prograde vector?

 

[Arrowstar]

I suspect onboard accelerometers in the form of inertia measurement units (IMUs) basically determine the orientation of the spacecraft with respect to the various accelerations (gravity) on it, data which can then be used to determine orientation wrt to velocity vector.

Maybe? :lol:

 

[Andy44]

Real-life spacecraft don't know unless they are told by the ground controllers.

This is how it works:

- A satellite is launched into an orbit which is determined by the launch vehicle's software.

- Using that orbit as a place to start, the ground controllers contact the satellite to command it and get telemetry from it. While this is happening, signals called "ranging pulses" are sent from the ground antenna, received by the vehicle's comm system, and retransmitted back to the ground. The time it takes for these pulses to get from the ground to the satellite and back again tells the ground controllers the range from the antenna to the vehicle. The antenna's az and el angles are also recorded.

-The range, az, and el is used as an input to a computer which runs orbit determination software. The result is a very precise orbit solution, which is then propagated into the future so that the next antenna knows where to find the vehicle when it's time to contact it again.

- The orbit solution may be transmitted to the vehicle's onboard computer (if it is fancy enough to have one) periodically. This tells the vehicle where it's at in inertial space and at what time. The velocity vector in inertial space is "prograde", and the software can do the math and figure out which attitude relative to the stars the vehicle must assume to point that way.

There are other ways, it might be possible to observe which stars rise and set at the Earth's limb at what times and do some sort of math to estimate an orbit.

In Orbiter, your spacecraft's orbit is automagically always known and thus Orbit MFD can show it at all times. The method of estimating an orbit is not simulated to avoid the more complicated and less glamorous side of spaceflight.

 

[SiberianTiger]

Which amazes me is how do they manage to steer the Cassini across Saturn subsystem and not to smash it into moons.

 

[Urwumpe]

Soyuz uses Ion flow sensors for telling where prograde is, which measure the motion of the Soyuz spacecraft relative to the ionized medium around it.

 

[Turbinator]

Soyuz uses Ion flow sensors for telling where prograde is, which measure the motion of the Soyuz spacecraft relative to the ionized medium around it.

Holly smokes wow, sounds something right out of Star Trek.

 

[Urwumpe]

Holly smokes wow, sounds something right out of Star Trek.

No, it was just reading the data of the third Russian satellite properly.

The Ion sensors have traditionally been pretty unreliable, but they are still used. A large part of the problems of Soyuz 1 in Orbit came from the Ion sensors.

 

[Andy44]

Which amazes me is how do they manage to steer the Cassini across Saturn subsystem and not to smash it into moons.

I've been trying to figure out how they do that. Doppler ranging will give you the velocity to or from the earth, but the entire Saturn system is just a dot in the sky, so I can't imagine they get any directional information from the antenna gimbals.

 

[mjessick]

I suspect onboard accelerometers in the form of inertia measurement units (IMUs) basically determine the orientation of the spacecraft with respect to the various accelerations (gravity) on it, data which can then be used to determine orientation wrt to velocity vector.

Maybe? :lol:

Accelerometers do not measure gravity. (Essentially all parts of the device are accelerated the same, so there is nothing to measure.)

An inertial space navigation algorithm has to add the acceleration of gravity into its state vector propagation using equations based on where it thinks it is. The greatest technical innovation embodied in the Delta Glider is not the 10 times better specific impulse, but the perfect navigation!

 

[Turbinator]

What kind of a education would you need, and what careers are there that deal with orbital mechanics and spacecraft, in this capacity? I find myself really enjoying completing complex orbit related tasks in Orbiter, and generally thinking about and researching spacecraft design. If I could make that in to a career, it would be amazing.




.

 

[Ark]

Huh, I always kinda wondered how they determined that. We clearly have a long way to go before we can zip around the solar system with impunity, at present it's hard to believe any crews would have a chance if all communication with the ground was cut off.

So, say the TLI burn for an Apollo mission was a few seconds short due to some faulty arithmetic in the planning. Would they have any warning in transit, or would they be clueless right up to the moment they get splattered across the face of the Moon?

 

[tblaxland]

I've been trying to figure out how they do that. Doppler ranging will give you the velocity to or from the earth, but the entire Saturn system is just a dot in the sky, so I can't imagine they get any directional information from the antenna gimbals.

They use Doppler for range and range/rate. That gives you a sphere (or a section thereof) constraining the spacecraft position to within about 0.5 m.

Optical navigation is used to determine the direction of the vector from Saturn (or one of its satellites) to the spacecraft. They do this by getting the spacecraft to take an image of Saturn or one of its satellites and determining what stars are in the background and their positions relative to the foreground object (they can use the Doppler pointing angles to get a rough first estimate).

The intersection of the Doppler sphere and the opnav vector gives you the position of the probe.

Cassini also has an onboard radar that can be used for relative navigation near the satellites and it has been used to more accurately determine the orbits, masses and gravity fields of the Saturnian system by combining its data with the Doppler/opnav data.

There is lots of good, if a little heavy, info in this report: Cassini OrbitDetermination Performance during Saturn Satellite Tour

 

[Andy44]

So, say the TLI burn for an Apollo mission was a few seconds short due to some faulty arithmetic in the planning. Would they have any warning in transit, or would they be clueless right up to the moment they get splattered across the face of the Moon?

Sure, they would know after the first orbit determination, which could happen very shortly after the TLI burn.

Figure to get a half-decent orbit solution they need 20 minutes of ranging data. They are in constant contact with the vehicle throughout TLI and afterwards, so they start collecting ranging a few minutes after the S-IVB cuts off, allowing enough time for the propulsion system to finish venting after shutdown. Once you've got the ranging data, it's a matter of how much time it takes to process it in the computer and spit out an orbit solution.

The mainframes they were using weren't lightening fast, but they were good enough for the job.

Of course, if they wait a little longer they can get a lot more ranging data and get a better solution. In real life, of course, they were getting new solutions on a regular basis and used them to plan the course corrections.

 

[Ark]

Sure, they would know after the first orbit determination, which could happen very shortly after the TLI burn.

Figure to get a half-decent orbit solution they need 20 minutes of ranging data. They are in constant contact with the vehicle throughout TLI and afterwards, so they start collecting ranging a few minutes after the S-IVB cuts off, allowing enough time for the propulsion system to finish venting after shutdown. Once you've got the ranging data, it's a matter of how much time it takes to process it in the computer and spit out an orbit solution.

The mainframes they were using weren't lightening fast, but they were good enough for the job.

Of course, if they wait a little longer they can get a lot more ranging data and get a better solution. In real life, of course, they were getting new solutions on a regular basis and used them to plan the course corrections.

Ah, cool. Collect data, transmit to ground, wait for computer to spit out an orbit plot. I had always kind of wondered if they were really just executing a well-planned dart throw.

So, what if communication with the ground was cut off? The Apollo computer didn't have much in the way of processing power, was there a procedure to collect the data and work out their solutions in flight?

 

[Quick_Nick]

Ah, cool. Collect data, transmit to ground, wait for computer to spit out an orbit plot. I had always kind of wondered if they were really just executing a well-planned dart throw.

So, what if communication with the ground was cut off? The Apollo computer didn't have much in the way of processing power, was there a procedure to collect the data and work out their solutions in flight?

Well even just because of the huge communication delay of such great distances, I assume they never leave the spacecraft in a dangerous trajectory.

 

[Ark]

Well even just because of the huge communication delay of such great distances, I assume they never leave the spacecraft in a dangerous trajectory.

Naturally. I was just wondering if they were capable of plotting their orbit and making course corrections without contact with the ground. Things can happen in-transit to affect the orbit, I'm sure Apollo 13's plot was measurably changed with the 02 tank blew and applied a sudden thrust.

 

[Zatnikitelman]

Um, it's not exactly a huge delay Earth-Moon, it's 1.34 seconds one way. So except for landing and taking off from the Moon, 3 seconds really isn't going to affect much in terms of getting data to them.

 

[tblaxland]

I was just wondering if they were capable of plotting their orbit and making course corrections without contact with the ground.

Yes, the Apollo Guidance Computer, in junction with the on-board sextant and IMU, could do orbit determination by doing a series of measurements of the positions of background stars relative to the Earth and the Moon (the Colossus software contained a rudimentary model of the Moons orbit). The AGC only had capability to perform targeting calculations for certain manoeuvres, but most importantly it could do return-to-Earth targeting. Read more in this MIT report (warning 8 MB pdf): http://ocw.mit.edu/NR/rdonlyres/Sci...B-4CFD-84D7-34CB82038893/0/1_4_9_mit_role.pdf

 

[Turbinator]

Don't the spacecraft also have a rudimentary internal navigation as a backup? Something along the lines of taking in to account all engine burn times, amounts of thrust applied, and in witch vector, and for how long. Also taking in to account star-tracking data, and anything else the spacecraft can measure. And determining from that what the orbit should be. And every time the spacecraft is in contact with earth this internal plot gets updated with exact data, to reset the error rate back to null, till the next contact? Especially for craft going to Saturn or the New Horizons going to Pluto and beyond. I assume that is what they use as a backup should communications fail, so that the spacecraft does not fly itself in to a planet until it can regain communication and receive new precise orbital data.



.

 

[Andy44]

Yes, the Apollo Guidance Computer, in junction with the on-board sextant and IMU, could do orbit determination by doing a series of measurements of the positions of background stars relative to the Earth and the Moon (the Colossus software contained a rudimentary model of the Moons orbit). The AGC only had capability to perform targeting calculations for certain manoeuvres, but most importantly it could do return-to-Earth targeting. Read more in this MIT report (warning 8 MB pdf): http://ocw.mit.edu/NR/rdonlyres/Sci...B-4CFD-84D7-34CB82038893/0/1_4_9_mit_role.pdf

That's very cool, I didn't know they could do that. Of course, if they lost all comm with the ground that would have be some major avionics failure. Given how much precision they always said was required for a safe reentry I always figured they were screwed without comm, but from your post I see that Apollo was much more capable fo independent navigation than I thought.