Wikipedia writes to say

As of 2009, about 19,000 pieces of debris larger than 5 cm (2.0 in) are tracked,[1] with another 300,000 pieces smaller than 1 cm below 200 km altitude

Control collision avoidance attempt to maintain a clear 2KM cube around the ISS. Whenever an object is expected the pass through this virtual cube, ISS has to manoeuvre. 2KM might sound a lot, but we're talking of a velocity of 7km/s - so it's not even a half-a-second away!

The orbit around Mars is probably not quite as crowded. On the other hand, there are 3 live spacecraft in orbit

  • MO
  • ME
  • MRO

There may also be other craft from older missions in Martian orbit.

With the imminent arrival of NASA MAVEN, and ISRO MOM, the number will increase this week.

Given the signal lag, is there any active collision avoidance system aboard any of these craft ? Is such a system needed at present?

  • 2
    $\begingroup$ You might find What risk management options are available for Mars orbiters and rovers to mitigate threat of impacting with Siding Spring cometary debris? relevant to what you're asking. A pizza box like the one the conjunction analysis is done for tracked debris many weeks in advance for ISS is unlikely to be needed, or indeed could be done for spacecraft around Mars and for potential collisions with other objects than the rest of the spacecraft there. But conjunction analysis for their own orbits is done... Well, this was a mouthful LOL :) $\endgroup$
    – TildalWave
    Sep 23, 2014 at 8:07
  • $\begingroup$ BTW it's not not even a half-a-second away if you do this analysis in advance. ISS has no way of maneuvering to avoid collisions in a split second. If the debris heading its way are too small to be tracked by ground radars and their orbit precisely determined, ... well, tough luck. And it does happen. That's why it's built to withstand such impacts. And sometimes, the crew will temporarily retreat to their escape vehicles during predicted flybys, just in case. All part of risk management. ;) $\endgroup$
    – TildalWave
    Sep 23, 2014 at 8:22
  • 1
    $\begingroup$ If they wanted to deliberately collide two of those spacecraft with each other on their different orbits, they wouldn't be able to do it, even where there are places the orbits could potentially cross. $\endgroup$
    – Mark Adler
    Sep 23, 2014 at 19:37
  • $\begingroup$ @MarkAdler Why? $\endgroup$
    – Everyone
    Sep 24, 2014 at 13:37
  • 1
    $\begingroup$ We can "only" control their orbits to within hundreds of meters. (Actually it's pretty amazing that we can do that well.) The spacecraft are only a few meters across. So even if we tried really hard to get them to collide at an orbit crossing, the probability of an actual impact would be vanishingly small. The Deep Impact probe succeeded at this sort of thing, but its target was much bigger. The only way to get them to collide reliably would be to put two of them in the same orbit and then rendezvous. However their orbits are all different enough that there's not enough fuel to do that. $\endgroup$
    – Mark Adler
    Sep 24, 2014 at 14:57

3 Answers 3


Given the signal lag, is there any active collision avoidance system aboard any of these craft?

No. For that matter, such an active collision avoidance system does not exist for the International Space Station, either. The Space Station moves when told to do so by the ground. The Space Station has no capability of detecting when a collision is imminent. At seven kilometers per second, detection of an imminent collision would mean an unavoidable collision.

The maneuvers by the ISS are based on predictions by ground-based assets with lots of computing power, which in turn are based on lots and lots of measurements made by other ground-based assets. We don't have radar stations positioned across the surface of Mars to make those measurements, and we don't have banks of blades and supercomputers dedicated to processing those non-existent measurements and solving the orbit determination problem.

The ISS is a very special asset. It is big, really really big, which makes the collision problem much more significant than anything else we have in space. It's cost is even bigger than it's size; it's loss would be catastrophic. That a collision could kill people makes the problem even bigger. The Three Letter Acronym organizations charged with keeping track of what's doing what in space are specially charged with protecting that singular asset.

  • $\begingroup$ I'm sorry but this is simply false: jpl.nasa.gov/news/news.php?feature=4572 $\endgroup$ Jun 2, 2015 at 18:02
  • 4
    $\begingroup$ @AleksanderLidtke - No, it isn't false. Read the article you yourself cited. All of the collision avoidance work is done by people on the surface of the Earth. There are no collision avoidance sensors on the spacecraft orbiting Mars. $\endgroup$ Jun 2, 2015 at 18:24
  • $\begingroup$ There are no collision avoidance sensors on pretty much any spacecraft, collision avoidance isn't done autonomously. So, in essence, the Mars orbiting spacecraft are operated exactly in the same way as Earth satellites, as far as collision screening is concerned. $\endgroup$ Jun 4, 2015 at 9:17

A few notes... The conjunction screening volume for ISS isn't a cube. It's a rectangular box which is longer in-track and smallest in the radial direction.

It kind of sounds like you're suggesting this volume is so that the spacecraft has time to maneuver. This isn't the case - any maneuver to mitigate a predicted conjunction would have to occur long before the event - at least half a rev is good, more is better.

The reason you have a screening volume larger then the satellite is because the satellite orbits have a significant amount of uncertainty, commonly referred to as the covariance. So even if you predict a 2 km miss between two objects, there is still a non-zero chance of collision because of the position uncertainty. The iridium/cosmos collision had a predicted miss distance from TLE data of on the order of 0.5 to 1.5 km, depending on which data you use.

This is also the reason why conjunction predictions, even for ISS are only done a few days to a week out. The position covariance grows with time so eventually it's to big to do useful analysis.

So back to Mars. Space is big - really really big. 5 objects in orbit around one planet should have a vanishingly small chance of colliding. With over 17K tracked objects around Earth, collisions are still a very very rare occurrence - 4 events to date.

The "space is big" argument isn't very satisfying however, because these objects aren't placed there randomly and there are orbit altitudes and inclinations which are more useful then others, so you would expect that objects would be concentrated there.

Back to covariance - does anybody know the accuracy to which we know the orbital positions of these probes? Sounds like a good follow-up question. If the covariance of the objects becomes to large, you really can't do effective collision avoidance analysis.

The best strategy may be to simply allow a sufficient altitude difference between the missions - which shouldn't be difficult given the population now or in the foreseeable future. But the risk really is quite low.

  • 2
    $\begingroup$ This does raise the interesting (to me) question: How many junk objects are there in Martian orbit? There are a small handful of dead orbiters and surely at least some pieces of debris e.g. from orbiter-lander separation events. One thing I am sure of is that we have no way of detecting or monitoring the orbits of any non-communicating objects in Martian orbit. Phobos and Deimos are just barely detectable by radar with Goldstone and Arecibo. $\endgroup$ Sep 23, 2014 at 19:23
  • $\begingroup$ @pericynthion: Great idea! That should make a good follow-up question! $\endgroup$
    – Everyone
    Sep 24, 2014 at 14:10

My initial thought was that, for a small number of orbiting craft, the orbits could be selected so as to not ever intersect, but some quick checking doesn't make it obvious that this is possible, since many of the orbiters have very eccentric, highly inclined orbits.

  • MRO, 93° inclination, 250km-316km altitude
  • MEx, 86.3°, 298km-10107km
  • MAVEN, 75°, 150km-6200km
  • MOM, 150°, 365km-80000km
  • Mars Odyssey, 93°, ~400km
  • Mars Global Surveyor, 93°, 378km

I'm not sure what the significance of the popular 93° inclination is; the three orbiters in that inclination all have very different orbital periods. It's theoretically possible that the periods of the orbits could be selected to harmonize, but I don't think it's likely; probably all the involved agencies watch each other's orbits and on the very unlikely chance a close approach is predicted, whoever's got the most maneuverable orbiter will nudge it.

  • $\begingroup$ Thanks for posting this - I didn't know they were that eccentric. $\endgroup$ Sep 23, 2014 at 20:22
  • 3
    $\begingroup$ Those popular 93° orbits are Mars Sun-Synchronous or Near Sun-Synchronous Polar Orbits. E.g. here's a short description of MRO's orbit. $\endgroup$
    – TildalWave
    Sep 23, 2014 at 23:22
  • $\begingroup$ Interesting; the physics of sun-synch orbits is a little beyond me, but this article might be informative for the more ambitious. trs-new.jpl.nasa.gov/dspace/bitstream/2014/6427/1/03-0272.pdf $\endgroup$ Sep 24, 2014 at 0:06
  • $\begingroup$ Ah -- Mark Adler has corrected a couple of these; the 93-degree orbits are less eccentric and diverse than I thought. $\endgroup$ Sep 25, 2014 at 0:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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