The race to destroy space garbage | Hacker News Reader

The race to destroy space garbage

bbc.co.uk

72 pointsCarolineW61 comments

I wonder if, in the future, something like 'Planetes' would become reality. Space janitors in some sense policing space junk in orbit be it operators controlling collection robots or actual human space-walk collectors themselves.

Collisions are the problem. When to large things collide they create hundreds of little things. So any space janitors, as large things themselves, would only increase the number of collisions and thereby exacerbate the problem.

The relative speeds of any two objects in low orbit means that physically catching much of anything is horribly inefficient. The DeltaV requirements for a robot grabbing one thing, then moving to grab another, means you are probably better off launching a separate robot for each target. Any complete answer must therefore involve some means of moving debris without physically touching it. Magnets and lasers are the only hope.

Disagree. What about a large elastic net?

They will not work on the scale of kinetic energies involved. Stuff with orbital velocities has so much energy that even metals start to behave like fluids, so an elastic net will be torn after the first impact.

This is not true because most satellites are in similar orbits, largely because it's easer going east due to earths rotation. Similarly, you could cheaply visit every satilite in geosynchronous orbit. Further, you can pick trash up in an order designed to minimize delta V, and use thing she like solar sails or ION drives.

Polar orbits are something of an exception, as are consolations designed for total coverage of the earth.

Similar orbits, similar inclinations, but still radically different relative inclinations. They arent all lined up in one big disk. They are in similar orbits spread out in time. So they are still aproaching each other at speed.

I agree there are a lot of useful orbits, but the most useful orbits like Geosynchronous orbit look like a thin donut/torus and have a lot of satellites all have the nearly the same vector just time shifted.

GPS satellites for example have basically a hot spare that can easily replace another satellite in each of the 6 planes. So, if you start out next to one GPS satellites there are 4 more you can easily reach. AKA visiting all 33 active ones would take 5 major burns, and lot's of tiny ones.

Sure, after a collision you can end up with a shotgun blast that would take a lot of effort to de orbit. But, having say 20 tugs that all deorbit the largest debris but getting them to LEO is a lot more manageable than you might think.

PS: The very tiny stuff is cleaned up by the solar wind over time and mostly not a problem in the first place.

Exactly. The tiny stuff can also be decelerated by solar wind and other random forces and fall back into the atmosphere.

The larger stuff on the other hand can cause serious impacts by transferring its momentum to the site of impact and causing damage to spacecraft.

An elastic net would let the small stuff zing around it while it could be deployed to pick up the larger stuff, by moving slightly fasted than geosynchronous orbit. Then it can actually toss them in various directions either slightly towards our atmpsphere or slightly away from Earth, to adjust its own course without expending fuel.

We could literally write a navigation program that would guide it on a path to clean up a lot of major debris this way without expending much fuel. It would only need its power for the tossing, which could be done with small explosions or mechanical pushes. The small stuff wouldn't alter its course much and the AI could correct it. Plus since it's in near earth orbit we could send updated routes to it.

Is anyone interested in writing code to route such a "space bot" by collecting and throwing stuff in various directions? We can simulate it in a 3D simulation with virtual random garbage of various sizes orbiting a planet.

Then you could present this sim to NASA or the Chinese space agency and make some big bucks :)

Geostationary orbits have no physical debris problems. There is far more real estate up there and relative speeds far less. Only a few low orbit groups share an exact orbit. Even sats launched on the same rocket drift into different planes over years unless they actively work to stay together.

To be clear, the near earth part of LEO gets cleared from simple drag though slower with increasing altitude. And even MEO is reasonably safe simply because we have yet to put a lot of junk there and there is more room.

So, even if the orbits where random you could simplify the problem by just sending more craft which would then need a smaller range of orbits. But, even within that model things are still far from random even though you will get a lot of drift over time.

Or you can try some technology that makes those huge delta-V viable, like magnetic sails.

Mag sails don't work in low orbits. The potential thrust they might generate is far lower than aerodynamic drag of such a large object. The magnetic field of the sail interacts with, collides with, charged particles of the upper atmosphere. Sails are only viable far away from any hint of atmosphere.