Basically via simulation these days. A course is simulated including burn times, when to burn, gravity, and a bunch of other stuff. The spacecraft then follows said course to get where it’s going. This stuff used to be done by hand, it’s all preplanned out regardless of what method you use. Computers are just way easier in today’s world.
Unless you literally mean how a spacecraft knows what direction it’s facing, in that case they orient relative to the sun and stars.
In order to navigate, you need to know your position and orientation (and velocity and angular velocity, but those can often be derived from position and orientation data taken over time). You then need to do some math to figure out how to get from your current position and orientation to your desired position and orientation. Then you need to actually do it. I’m going to gloss over step two, but here are a number of devices that are used to determine and control position and orientation.
Depending on where you are in space, you may need to use a variety of different methods to determine your position and orientation.
If you are close enough to Earth, you can use:
* GPS (position) (works just like it does on Earth)
* Ground tracking (position) (shoot radar off of the satellite and see how far it is from the radar station)
* Earth sensor (rotation, rough) (If that direction is bright, Earth is probably in that direction)
* Terrain relative navigation (rotation and position) (look at Earth to see where you are) (Frequently used when landing on the Moon, not as much on Earth)
* Magnetometer (rotational velocity, rotation if you are edgy, maybe position if you are very edgy) (Earth has a magnetic field and you can measure it to figure out how fast you are spinning and what part of it you are in) (usable near other planets with a magnetic field)
* Radar/Lidar altimetry (position, possibly rotation but not usually) (see how far from Earth you are by bouncing radar or light off of it) (not normally used on Earth, common when landing on the Moon)
If you are not close to Earth, or even if you are close to Earth, you can use:
* Gyroscopes (rotation) (spinny thing tends to stay spinning without changing spin direction, if your spacecraft spins you can check the spinny thing to see how much you’ve spun by)
* Signal timing (position) (determine how long a signal takes to reach a spacecraft, gives distance)
* Star tracking (rotation) (look at the stars and do a bunch of fancy math to see how you are rotated, the stars are so far away that they don’t vary in position too much, so you can use them as a fixed background that doesn’t rotate, a bit like a gyroscope)
* Sun and/or planet sensing (rotation and position if you know the time) (Star tracking but for planets. Because planets move, if you know the time, you can do the math on their orbits and get your position within the solar system)
* Inertial guidance (rotation and position, not typically accurate over long timescales) (really accurate accelerometers (linear and angular) that can determine how much you have moved relative to the orientation you started at, so you can know where you are relative to your starting position)
* Sun sensors (rotation) (small light sensors around the spacecraft, can determine rotation relative to the sun by comparing how bright each one is)
Most spacecraft use a combination of these methods rather than relying on a single method as they all have their advantages and disadvantages. There are doubtless a few I have forgotten and/or don’t know about, and a number of theoretical systems that could work but have never been tried for one reason or another.
Ran out of characters, see part 2.
For a spacecraft going outside of Earth orbit, most of the navigation happens at launch and very early on, with only minor corrections later on. The problem you run into is that every kilogram of stuff you want to put into space is really expensive and requires a lot of fuel, so if you’re trying to put a whole propulsion system *and* fuel for it you need a way bigger rocket to launch it all. The Apollo missions had this challenge, and while it’s feasible it’s something you’d rather avoid.
So how do you navigate at launch? Well, the great news is that planets, asteroids, moons, the sun, etc. all move very predictably, so we can model a route that our spacecraft will take for the next 5, 10, 20, 30 years all before even launching it. At that point, we wait for the launch window when we’ll be starting on that route, send the rocket up, and effectively just chuck the craft into space along that route. After that point, we let gravity do all the work for us. Minor course corrections might be done due to impacts with small things or momentum imparted from solar storms, the force resulting from reangling instruments, etc, and those can be done with small boosters we put on the craft along with what fuel we do send up. Those minor corrections can have really big impacts when you’re talking about traveling millions of miles over years of service, but for the most part once the thing is disengaged from the boosters and the payload is traveling freely, the bulk of the navigation is done.
Latest Answers