Assuming such a space station needed to have its central hub not used for docking spacecraft, would it be plausible for a spacecraft to dock on the outer edge of a rotating wheel if the ship approached the station on a tangent to the outer edge, in the direction of rotation? I'm assuming a drone ship, of equal mass, would simultaneously dock on the opposite side of the station to maintain the balance of the mass of the station.

  • 1
    $\begingroup$ Why the drone ship? Are you thinking the station is like a centrifuge? $\endgroup$
    – Willk
    May 29, 2019 at 2:41
  • 1
    $\begingroup$ @Willk Yes, I'm thinking of the station as a centrifuge. $\endgroup$
    – Bob516
    May 29, 2019 at 2:44
  • 4
    $\begingroup$ "Well, captain, we can go for the high-risk-of-catastrophic-failure-for-little-reward tangent dock. Or we can just dock easily and safely at the hub with everybody else. The ship owners and crew have entrusted you to complete our mission safely, on time, and on budget. What are your orders?" $\endgroup$
    – user535733
    May 29, 2019 at 2:53
  • 1
    $\begingroup$ This possibility and its drawbacks are covered in this more general question What is the best design for docking onto a rotating space-station?. Possible duplicate? $\endgroup$ May 29, 2019 at 7:41
  • $\begingroup$ Seems like a lot of trouble to go to when you have a nearly infinite amount of hub you can build onto $\endgroup$
    – John
    May 29, 2019 at 15:47

5 Answers 5


Can you launch an ICBM horizontally?

Sure. Why would you want to? (The Hunt for Red October)

You absolutely can dock on the outer rim. It will be the biggest pain in the tuckus your pilots have ever seen, but yup, it can be done.

There are (at least) two reasons why the central hub is preferred.

  1. You only have to spin the ship on one axis, if you must spin it at all.

  2. You can design the hub so that the docking hub doesn't spin, which is even simpler.

But hitting a docking port on the outer rim means the ship must move on at least two axes. That's what it takes to move in an arc: two axes. This is because you can't actually move up to a point tangent on the outer rim and dock.


Because docking takes some time. Even if you're trying to capture the ship with some cool clamps. The ship is moving in one direction (a straight line) and the outer rim in another (an arc). The "docking time" isn't just a second... it's a minuscule fraction of a second — and if you don't get it right the first time, things (usually the ship) get ripped apart.

So, theoretically, yes, you can dock a ship via the outer rim. It's not recommended.

  • 3
    $\begingroup$ Given time, money and an unhealthy dose of masochism, everything is possible :D $\endgroup$ May 29, 2019 at 2:57
  • $\begingroup$ Could be doable if you had a docking bay that was a big scoop and you timed it juuuust right. Then all you’d have to worry about is slamming into the back wall of the docking bay! $\endgroup$
    – Joe Bloggs
    May 29, 2019 at 10:11
  • $\begingroup$ And the station itself will be at a high risk itself. $\endgroup$
    – NomadMaker
    Aug 20, 2020 at 14:08

As others have explained, It is quite a pain and a lot of risk with docking ports that resemble the actual ones we use today on the ISS and elsewhere.

But I think it may be possible with a tow-capable tether. A somewhat high-tech version of how surface ships are moored.

  1. The docking spacecraft approaches and only roughly matches the rotating docking port. This requires the continuous expenditure of delta-v mass but hey, we are in sci-fi land.

  2. One of the spacecraft (doesn't matter which one) reels out out a couple (or more) of strong tether cables, each piloted by a remote-controlled drone thruster. These maneuver and attach themselves to the hardened attachment points on the other vessel. The pilot thrusters and attachment points are hardened enough to handle rough contact, and the thrusters are nimble enough to be very maneuverable. (they don't have to accelerate their delta-v mass. It's fed through the tether)

  3. Once the tethers are secure, they are slowly reeled in. The docking ship slowly lets go of the thrust as it approaches the station.

  4. As the tethers go taut, the drag will slow the station's rotation down slightly. The station may fire thrusters to counteract this, or if this is tolerable, the docking ship may fire its own thrusters for the same purpose once it is securely anchored.

  5. Once the ship is close enough, robotic clamps engage and secure the two spacecraft together.

  6. Only when the spacecraft are mechanically secure, extendable airlock ports deploy and internally connect the two vessels together.

As you can see, this is way more convoluted than hub docking. But hub docking can only properly accommodate two ships, while with rim docking, docking capacity is limited only by the size and surface area of your station.

The two hub docking ports may still accommodate large ships that would destabilize the station if they docked on the rim.

It may be an acceptable compromise for a station that serves (ironically) as a transport hub. :D


Orbit tethering

This is another, probably more economical and easier method for rim docking that I brewed up after I posted the answer above.

Again this involves tethers and pilot thrusters as above, but requires much less delta-v for the docking maneuver, is much simpler and more stable.

  1. For this approach, the spaceship parks at a safe and stable point above the rim, aligning its main thrusters with a hypothetical tangent of a circle concentric with the station axis.
  2. Tethers are extended from the two sides of the station hub, attached to bearings whose rotation has been neutralized (ie. they are stationary from the docking ship's reference frame)
  3. As above, the tethers are attached to the two sides of the docking ship.
  4. Using its main thrusters, the docking ship accelerates prograde to the same linear velocity that the rim docking port is traveling in. The tethers ensure that the ship remains at the same radial distance. This somewhat emulates a stable orbit (the tension on the tether replaces gravitational force)
  5. The station slowly reels the ship in. Rotational inertia is preserved, so that the angular velocity of the ship's artificial orbit matches up with the angular velocity of the rim port. The station can sync the reel-in speed to align the ship and the port. (Initially, the port will be rotating faster than the ship, and reeling can be timed so that the ship and the port line up by the time it's done).
  6. When the ship and port are aligned, locks are engaged, and the tethers are released for other docking duties.
  • If somehow a vessel misses the faster rotating port, the tethers can be loosened, slowing the ship down. This allows the port to go around faster, giving another opportunity to sync rotations. Smaller adjustments can be done by prograde burns alone. (Retrograde may be dangerous at low altitude)
  • Having four tethers that can be independently reeled allows for very precise attitude adjustments so that the first contact happens without a big jerk.
  • The ship is going to experience some artificial gravity. It makes sense to align the pilots' axes with the radial vector, heads toward the center, so that what they experience feels somewhat natural. This makes the docking point "above" from their perspective.
  • Assuming that the mass difference between the ships and the station is at least a couple of orders of magnitude, it may be possible to dock on the sidewalls of the rim as well as the outer edge. There may even be port ratings where the lightweight ships dock on the sidewalls, middleweight ships on the outer edge, and at most two heavyweights directly on the hub.


That would make no sense as it's too complicated and too many things would go wrong.

The station would have a self balancing system which would operate by pumping water to holding tanks around the rim.

As the ship docks, water would be pumped from the area where the ship was to the area opposite the ship.

  • $\begingroup$ The ship docking will be instantaneous, but the water ballast transfer will take significant time. And during that ballast transfer the rotation of the station will be affected. Even if the station can withstand the stress, the inhabitants are going to be seasick every time a ship docks or undocks. $\endgroup$ May 29, 2019 at 12:40
  • $\begingroup$ Can't possibly be instantaneous. The ship is going to have to match rotation using its own jets. It would taper off it's jets as the station corrects the imbalance. I'd also like to point out this is why you'd never dock on the outer ring of the station. It's the worst spot to dock. $\endgroup$
    – Thorne
    May 29, 2019 at 23:07

Yes, but not recommended

It will be a jerk, though. The space station has one or more circular hoops mounted on it's outer shell, preferably with some distance. From each hoop, several docking apparatuses hang, like a crane from a girder.

The space ship flies tangentiall to the hoop, it's docking interface pointing upward and at a speed equal to (or close) the speed of the hoop as a whole.

The docking apparatus adjusts it's position along the hoop and in the right moment snatches the docking interface of the ship with a powerful magnet. This will be experienced as a strong jerk by the ship.

Then, once the ship is grappled, the actual docking with matching airlock interfaces etc. is done between ship and docking apparatus.

And then, at last, the docking apparatus drives near an airlock to the station and docks. Can't have a moving apparatus that's permantently connected to the station (or you could, with flexible walkway and a severly limited range for the apparatus).


  • the ship has to sustain it's own weight from the docking interface under a bit more than the stations gravity, same for the docking apparatus and hoop

  • strong jerk

Would the maneuvering be easier or harder than Elite-style docking? With a stanford torus (the least practical shape for this kind of docking), the outer rim moves at about 93 m/s, this is the speed the ship needs to have. This also means that the timeframe for the grappling maneuver is pretty short and depends on how far out the docking apparatus can reach (physically and via it's magnets). But I'd assume an order of magnitude of 0.1s.

With circumferential docking, the ship has to be move along the correct axis at exactly the correct speed, the docking apparatus has to adapt accordingly. The docking course is almost a near miss, so it would be trivial to change away from a collision course if need be.

With elite style docking the axis has to be correct too, with the rotation and phase as additional variables. The docking approach can in theory be arbitrarily slow, so collision prevention shouldn't be too hard either. I's assume something similar to the docking apparatus would be used here, too, just inside.

In summary, elite style docking is probably the better approach but a station may employ both for easier traffic management.



...and it depends on what you mean by "docking".

SpaceX has gotten a lot of press for being the first to do this with their booster rockets (as opposed to others historically dropping their boosters into the Main Seawater Retention Tank). Various crewed spacecraft have been doing it. Freefall (webcomic) recently did it, so at least someone else has theorized that it's possible. Halo has certainly done it.

You almost surely have to approach this the way Freefall and Halo did, and treat it not as a "docking" (which would be pretty insane) and more as a "landing". The moment you're "inside" the station, you are essentially having to hover in its pseudogravity in order to avoid crashing... but as noted, we do know how to land in similar conditions. Obviously, the larger the station (or more to the point, the slower it is rotating), the easier this will be.

The trick is likely going to be that the station needs to have enough mass (relative to the ship "docking", by which we really mean "landing") that the change in rotational inertia is close to negligible.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .