4
$\begingroup$

Use of sensors and lasers seems obvious, but what all goes into the system that achieves this? What does the process entail? Example includes SpaceX's recent autonomous docking to ISS of Crew Dragon.

$\endgroup$
  • $\begingroup$ There are many, many books on this subject, and even more articles in peer reviewed technical journals on this subject. You have unknowingly asked us to write a book. $\endgroup$ – David Hammen Jun 9 at 10:45
2
$\begingroup$

The Russians have used a docking system known as Kurs for quite some time. Progress, Soyuz, and the ATV (From ESA, using equipment from Russia) dock all the time in automated mode using Kurs.

SpaceX developed its own docking/approach hardware/software and initially tested it on a Shuttle mission before (to pretend to dock along with the shuttle) before using it on the Dragon Cargo capsules.

| improve this answer | |
$\endgroup$
2
$\begingroup$

Once line of sight is established to the selected docking port, the International Docking System Standard (https://www.internationaldockingstandard.com/) in section 3.5 defines three visual targets that can be used to measure the relative orientation between the spacecraft and the port.

There are 3 perimeter reflector targets (PRT) which are retro-reflectors that can be used from longer distances. Once closer there are peripheral docking targets (PDT) and a centerline docking target (CDT). They act similar to iron sights on a firearm.

The targets can be seen on this picture from the IDA-2 prior to installation on the ISS: https://commons.wikimedia.org/wiki/File:IDA-2_upright.jpg.

The PRT are the tiny black features standing a bit of from the port, there are two on the top (one is in a blue clamp) and there is one on the right. The PDT is on the bottom right, you can recognize the circles standing out. The CDT is fixed on the canvas behind the IDA, it is partially obscured in this picture.

Edited to add: While I haven't found definite information on Crew Dragon, on STS-127, STS-129 and STS-133 SpaceX tested the DragonEye LIDAR system which was designed for the old Dragon capsule to locate the retro-reflectors on the station. For the closer approach there is a camera at the center of Crew Dragons docking port, directly opposite to the CDT on the Station.

You can see pictures of the camera on the official NASA video at 27:02:

| improve this answer | |
$\endgroup$
  • $\begingroup$ This is a good discussion of the targets. With some added information about what sensors are used with the targets, it could be a great answer to the question. $\endgroup$ – Organic Marble Jun 7 at 17:30
  • 1
    $\begingroup$ @OrganicMarble ok, I added information on the DragonEye LIDAR and the centerline camera. $\endgroup$ – David-H-K Jun 8 at 20:30
  • $\begingroup$ Great stuff! There is some info on Dragoneye here nasa.gov/pdf/358018main_sts127_press_kit.pdf $\endgroup$ – Organic Marble Jun 8 at 20:46
-1
$\begingroup$

It's actually quite simple. A spaceship is really have the simplest model of dynamics and control: you have independent control to all 6 degrees of freedom and they are all perfectly linear and newtonian, and you have almost no disturbance or interference. Comparably cars, airplanes and helicopters are all much more complicated (more and more degrees of freedom are coupled together and environment could disturb you badly).

PID controller

Basically,

P) the further away, the harder you push (stop pushing when you have reached target)

I) Gradually push harder and harder, when you are far, but be careful not to overshoot (gradually reduce force or push backwards when you are close)

D) watch your speed.

Human astronauts basically does the same thing in manual mode: like this

| improve this answer | |
$\endgroup$
  • $\begingroup$ Relative motion in low-Earth orbit, which the ISS is, is well-approximated by the Clohessy-Wiltshire equations, which are nonlinear. en.wikipedia.org/wiki/Clohessy%E2%80%93Wiltshire_equations $\endgroup$ – Erin Anne Jun 9 at 4:37
  • $\begingroup$ @ErinAnne - Your comment is incorrect in a number of ways. One, the CW equations are linear. They are a linearization of the underlying dynamics that indeed are nonlinear. Two, using a linearization of the underlying nonlinear dynamics is a very common practice in control theory. Three, the linearization does not say how to control the spacecraft's motion. That's what control theory does. (continued) $\endgroup$ – David Hammen Jun 9 at 9:57
  • $\begingroup$ That said, spacecraft typically do not use PID control for docking because the vernier thrusters used for docking are either on or off rather than throttleable, they point in fixed directions, and there's typically a lot of cross coupling between attitude and translation in terms of how thrusters affect vehicle motion. $\endgroup$ – David Hammen Jun 9 at 9:58
  • $\begingroup$ @user3528438 - Spacecraft docking does not use PID control. As a starter, you might want to google (1) "spacecraft jet select logic", (2) "phase plane control", (3) "H-infinity control", (4) "linear–quadratic Gaussian control". $\endgroup$ – David Hammen Jun 9 at 10:15
  • $\begingroup$ The translation of hand controller inputs to jet commands is nontrivial and imperfect. If you watched the SpaceX Demo-2 docking, there was a point where the astronauts took the vehicle into manual control mode. The astronauts noted that control in Y was a bit sloppy, but they were expecting that sloppiness because that is exactly what simulations showed. If spacecraft were perfectly spherical cows, there would be six sets of five thrusters (or, for redundancy, six sets of fifteen thrusters) so as to give uncoupled translational in and rotational control about each primary direction. $\endgroup$ – David Hammen Jun 9 at 10:27
-1
$\begingroup$

Everything above is true, but it’s harder than it sounds. Read up on Gemini 4. During that mission, the astronauts tried to fly towards the expended titan booster as a step towards rendezvous. It was a mess. Nothing worked as expected.

Orbital mechanics are pretty counter-intuitive. Your altitude determines your speed, so sometime speeding up means vectoring down, etc. They figured it out by the end of Gemini.

So the astronauts tell the ship to go “that way” and the computers fire thrusters in a variety of directions to make “that way” occur.

| improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.