• A spacecraft is travelling between planets. It accelerates at a constant rate for the first half of the trip, and then decelerates on the other half. Throughout the journey, aside from the switch in direction in the middle, the occupants of the ship experience a comfortable consistent gravity. Longer trips are maybe pieced together from several pairs of acceleration/deceleration legs, such that at no point does the spacecraft have to deal with particles hurtling at them at relativistic speeds.

  • Instead of dropping the entire ship down a gravity well and hauling it back out, smaller ships (better equipped for reentry and other aerodynamic shenanigans) are used to transport passengers and cargo between spaceship and planet. The ship waiting in orbit moves a little faster than a free object in the same orbit otherwise would, and provides the necessary extra centripetal force by continuously firing its thrusters in the direction away from the planet below. Crew waiting aboard the ship experience artificial gravity with “down” pointing towards space. (Or vice versa, with a slow ship and propulsion away from the planet instead of into it, and artificial gravity in the opposite direction.)

  • A space station with artificial gravity produced in the aforementioned fashion. A space elevator connects it with another station in a higher orbit, which experiences weightlessness because it is orbiting with the same period as the “bottom” station. The “top” station provides a controlled weightless environment, perhaps for all kinds of zero-gee science, while a permanent/long-term crew are rotated through the “bottom” station so that they experience gravity often enough to keep their health from crashing. Traffic between the station and the planet below carries cargo (food, propellant, other necessities) exclusively.

The idea is to remove the problem of adapting to weightlessness, simply by having every tin can in space that has people in it move in a way such that gravity is experienced nearly all of the time. This question is about the tradeoff between the cost of propulsion (propellant, and the maintenance of engines that are kept on 24/7) and the convenience, and other benefits, of having gravity.

Because burning expensive chemical fuel just so things don’t float away some of the time is extravagant to an appalling degree (and simply isn't sustainable anyway), we might suppose fusion (or some other method of getting a lot of cheap energy from very little mass) is available.

edit: the intention of the previous paragraph, and the wording of the following question, had been to indicate that the problem of generating the sustained high thrust required for the implementation of these schemes is to be taken as issue that can be circumvented. They were written based on the assumption that access to cheap energy implies the capability to generate those thrusts; as the answers from Rekesoft and Firedrake below have pointed out, that assumption is not true.

Nevertheless, I am still interested in being informed of other problems that are relevant to the scenarios described, now supposing that some sort of magic space drive is employed.

What other obstacles must be overcome to have widespread use of artificial gravity be practically favourable? What other important things am I overlooking?

Some more things I’ve considered:

  • I don’t think I’ve done sufficient research on ways to alleviate the detrimental effects of microgravity on human bodies to judge this, but I suspect just having some gravity (say several tenths of a gee) might already be very helpful. In that case, doing some simple maths and taking data from some wiki pages, It seems like having around 0.2 gees would be, considering the thrust required kilogram by kilogram, about four times as expensive as keeping an airliner in cruising flight. I have no idea how economically sound this would be, but feel like it may not seem too unreasonable if propulsion is sufficiently cheap.

  • Possibly relevant is this question, in which artificial gravity is achieved through the same mechanism in a ring habitat, except the load is supported by structural integrity instead of propulsion.

  • $\begingroup$ On longer travels, you don't need to divide it in several accelerate-decelerate legs. There isn't much of a velocity limit in space, so it's better to have only one accelerate-decelerate leg: fuel and propellant consumption is the same, and travel time is much shorter. Velocity limit would be when you reach a big portion of light speed, at which point space dust and gas cause can endanger even heavy forward shields. But this is far beyond what would be attained in such trips, unless you plan to have interstellar travel. $\endgroup$
    – Eth
    Commented Sep 11, 2017 at 16:22
  • $\begingroup$ I'm not even sure what you're trying to describe with your third bullet (blame my lack of imagination) but for your two cases involving orbit I think you're overcomplicating things. Imagine a spacecraft that can thrust over long durations at 1g (your premise). Don't even go into orbit, just hang out in space and thrust away from the planet. For example, at the ISS gravity is roughly 0.9 g's (astronauts just don't feel it as they are in free fall). If the station were not in orbit but rather constantly thrusting away from Earth, its occupants would feel 0.9 g's towards Earth. $\endgroup$
    – ben
    Commented Mar 2, 2019 at 21:08

4 Answers 4


The main advantage this system has it's not the gravity it provides, but the shortening of travel time. A constant acceleration, even if weak, is the way to achieve ultra-high speeds, as long as your engines can deliver for so much time. A few days accelerating at 1G would get you travelling at an appreciable fraction of speed light, thus shortening the travel to the point that microgravity effects are negligible - I'm supposing solar system travels, not starships.

I don't think there's any kind of technology that would make it economically feasible, though.


Yes, this is acceptable. It is basically the solution used in the Expanse novels by James S.A. Corey, as well as a few others like Mike Kupari. Expanse ships have incredibly efficient engines so they can cruise around under thrust, usually at .3g or so, but able to go up to 3-4 g or higher for short periods. In orbit you would still be in microgravity unless docked to a spinning station however. This makes sense, there are only so many locations in the solar system so a station of some sort would be at almost anywhere worth going to.

Prolonged zero-gee is very detrimental to humans and this gets around that problem (assuming the characters transition from zero-gee to higher gee regularly, if you stay in zero-gee the changes may not be as dangerous). There are acute changes when going to zero-gee (vertigo, sinus congestion, etc) that this solution doesn't address, but those can be (and almost always are) comfortably overlooked.

It is virtually impossible to carry enough remass to stay under thrust for very long, but there are torchship concepts that require little physics handwaving and still let ships use thrust for most, if not all, of a planet to planet trip. Or just say that the ship collects and uses dark matter for remass (80% of the universe!) and thrust around to your hearts content :)

  • 1
    $\begingroup$ If you want to use the "runs on efficiency" drives from the Expanse but be slightly more realistic, you can add big propellant tanks and giant red-hot radiators. Hard-SF-minded people will assume your craft use some undefined futuristic high-end nuclear drive instead of the near-magic of the Epstein drive. Other people will think your spacecrafts with their giant red-hot radiators look pretty badass. And as always, make sure to not turn the drive on near anything inhabited (or with unhardened electronics, for that matter). $\endgroup$
    – Eth
    Commented Sep 11, 2017 at 16:30
  • 2
    $\begingroup$ @Eth spot on. I should have mentioned the astounding www.projectrho.com Atomic Rockets site, since it answers this (and almost all other science based) rocket questions. $\endgroup$
    – Jason K
    Commented Sep 11, 2017 at 16:38

Thrust is not the answer: there are no current or plausible methods for producing sustained thrust high enough to be biologically useful. (Everything is either high thrust and low specific impulse, like chemical rockets, or low thrust and high specific impulse, like ion drives and plausible ideas for fusion.)

You might do better either to have a ring (free floating or part of the spacecraft) or to tether two spacecraft together and spin them round their common centre of mass. Then you need to build fairly large spacecraft by current standards (or use a long tether), because a fast rotation is disorientating to the inner ear, and the larger the craft the slower it can be to get the same acceleration. Also: spin complicates manoeuvres (gyroscopic effects; having two counter-rotating rings can help a bit); spin complicates docking; a spin section requires very reliable bearings.

If you do deploy a magic space drive that can produce sustained high thrust fairly cheaply, you need a reason why planets still exist (since anyone with a drive can build a kinetic weapon that'll destroy it); but also rendezvous gets complicated again, because the two craft have to match their non-zero thrust before they can start manoeuvres to dock.

Another complication: sometimes the drive will fail, which is exactly when you don't want your ship crews to be in a new and unexpected situation.

Another complication: zero-g manufacture makes assembling large spacecraft much easier. (And for some materials may be necessary.)

  • $\begingroup$ For a while I thought matching velocity and acceleration wouldn’t be that much harder in principle. I then realised that spacecraft are likely to have one big engine providing most of the thrust, while several smaller ones rotate the ship to point it in the right direction. Maneuvers don’t just become more mathematically involved - for some configurations, docking might be impossible unless the spacecraft fallback to travelling together at constant velocity (for example, two ships with airlocks with “this way up” signs that only match when their main thrusters point in opposite directions). $\endgroup$
    – user42460
    Commented Sep 17, 2017 at 13:51
  • $\begingroup$ I would argue that a Zubrin-type NSWR type drive is at least somewhat plausible, for a given level of plausible (possible, significant engineering challenges and capable of irradiating a landing site for the next few hundred thousand years, but still plausible). Otherwise, good answer. $\endgroup$
    – Gryphon
    Commented Feb 11, 2019 at 21:36

Kangaroo Comet

A ship in an orbital path creates the condition of weightlessness by falling toward the centre of gravity at the same speed it overtakes it, so to speak. In low earth orbit this is practical.

In interplanetary trajectories the nearest centre of gravity will be far enough away that the ship can, for a period, intentionally fall inward toward it at a velocity faster than that at which it overtakes it. 'Up' and 'down' then are created along a line somewhere between the lines of the two forces being considered, giving a degree of gravity on board. When it comes time to correct the ships heading, the crew compartment swivels in its gimbal and the ship accelerates away from the centre of gravity.
For shorthand, one might characterise this as 'zig zag' or 'kangaroo hop' gravity, as the actual path of the ship traces a line similar to an epicycle around the theoretical comet path, (and on that same plane).
One fantasy element which needs yet to be made into physics is how to power the manouvre.
The ship zig zags around a comet-like path. As potential is consumed during falling, the velocity of the ship increases, and as it zags away from the undershot path it will slow in a ballistic path. Power will be needed to correct the headings for each zig and zag, and for reorienting the crew compartment. Artificial gravity will not be constant, but at least there will be an up and down.


You must log in to answer this question.