It is very rare to
perform collision
avoidance manoeuvres
with active
satellites. The vast
majority of ESA
avoidance manoeuvres
are the result of
dead satellites or
fragments from
previous collisions.
[ ... ]
As the number of
satellites in orbit
increases, due to
'mega constellations'
such as #Starlink
comprising hundreds
or even thousands of
satellites, today's
'manual' collision
avoidance process
will become
impossible...
EDIT: Said comment is now one full screen tall due to manual insertion of linewraps, and is very wasteful of space on desktop and landscape layouts: https://i.imgur.com/H15eCXo.png
It so weird that there is no incentive to fix many usability shortcomings like this one and too tiny UI elements on mobile etc. - all of that despite multiple complaints over years..
Mobile works fine for me and the UI elements as shown honor my text size setting accurately. Using code formatting for non-code results in broken formatting for non-code. This should come as no surprise to anyone. Indents aren’t worth a semantic violation. Use “> ” prefixes like classic email, and optionally italic asterisks if that’s desirable.
There was a post on HN yesterday thanking the operators for never updating the site. While I mostly agree with that sentiment, this code formatting BS is an outlier and very much needs to be addressed. Either take the feature away altogether, or just make it work well on mobile. But this middle ground where HN visitors inadvertently ruin their contributions for mobile users is bad.
It is very rare to perform collision avoidance manoeuvres with active satellites. The vast majority of ESA avoidance manoeuvres are the result of dead satellites or fragments from previous collisions. [ ... ]
As the number of satellites in orbit increases, due to 'mega constellations' such as #Starlink comprising hundreds or even thousands of satellites, today's 'manual' collision avoidance process will become impossible...
For anyone not familiar with aerospace terminology: LEO is not the name of the satellite but just means that it is a low flying satellite. It stands for Low Earth Orbit.
How is it the case that StarLink didn't realize they were launching into orbits that would intersect with others?
I find it foreboding that the number of intersections with space debris and live constellations is to the point where ESA is moving towards non-deterministic algorithms to hopefully avoid the worst. How will Musk ever leave this planet if we make a shell of debris around ourselves?
Since all these orbital parameters change over time due to atmospheric drag, solar pressure, gravity irregularities, etc, organizations like NORAD constantly monitor and determine the orbital parameters of every object they can find circling the Earth.
So, they watch for when the likelihood that one satellite will come within some small distance (say, 5km) of another is high. For example, if the odds that they come that close are say, 75% in the next 5 orbits, the operators are notified and alter their courses slightly to maintain separation.
Even manned spacecraft such as the ISS will perform separation maneuvers to keep themselves clear of objects that may come too close.
I think the issue is that people have some vague affinity for the way space travel was depicted in old science fiction, where you could blast off and manually control everything with possibly the assistance of a slide rule.
And it's slightly discomfiting to realize that just traffic control is so complex that it needs modern computers to track and react. Space is supposed to be vast and empty! And it is, but not in LEO.
Clearly we need to invent ESA a Space-debris Hunting semi-Automomous Reconnaissance Craft, with a solar-powered infrared heating beam to shatter the frozen space material into ever-smaller bits. Or, as I like to call this class of spacecraft -
I wonder if AI is really the right tool to automate the process. Couldn’t it be done more robustly with a simple algorithm that takes todays collision probability models as input?
How much thrust? How high should it go? And on what information? Data sources will become more complex, risks need to be weighed. I'm not sure 'AI' is the right answer as most (all?) of AI is probably today just ML or more often simple BS - but certainly what ESA & co need won't be a simple one-variable algorithm.
I'm actually wondering if there's not a need to communicate more between all the bodies having satellites in space - just like autonomous cars, if both make an evasive maneuver they might cancel each other out. So there needs to be some kind of communication to avoid this.
- most popular orbits are also most crowded and dangerous ones, there's only so much you can do to reduce the probability of collision
- there are also uncontrolled debris, non-tracked objects, military satellites with unpublished orbits etc
- just like with automating road vehicle traffic, it would require very high degree of centralization, which is never going to happen due to various reasons
- the propellant is limited so you also want to reduce the amount of avoidance maneuvers to a minimum
So no, statistical models (which you probably mean by AI) won't improve the situation much. What matters is:
- increasing the orbit determination accuracy and coverage with better instruments and commercial tracking networks; NORAD TLEs alone are not precise enough for reliable collision prediction
- better space weather prediction
- improving the coordination between the operators
- deployable decelerators for dead satellites in lower orbits, also possibly debris removal if the problem becomes really serious
i.e. stuff most operators already do, except for the debris removal
Predictions based on TLEs aren't precise enough for satellite collisions. TLE is a solution compiled from many measurements, and what one wants for very close passes are raw measurements. It's a bit like JPEG vs RAW in photography.
Regarding that particular collision, IIRC human factor was involved a lot. Iridium has been warned well in advance, but hesitated to make the decision since their models gave lower probabilities.
Of course, inside the article it’s explained that SpaceX didn’t “refuse” to move the satellite, they were politely asked to do so and declined for reasons the author is unaware of. Possibly because they couldn’t move the satellite or had a higher collision tolerance.
The thing is that satellites are planned with a lifespan which can include a fixed amount of fuel for adjustments like this. It's not too hard to attach a dollar amount here.
I'm as big a fan of SpaceX as you'll find but considering they were the one entering an orbit they didn't previously occupy it seems like a pretty lame move on SpaceX's part.
This implies that un has jurisdiction over all orbits, which it can't really enforce. Also, good luck getting any kind of int'l tax passed; no one wants to give up that kind of sovereignty. Were I a company, I'd just wait until office changed and whatever treaty this was got torn up to put up my satellites. How could the un enforce this? At the end of the day, policies are enforced with the threat of force in response to non-compliance; fact is, un has precious little of that. The only viable enforcement mechanism would be enforcement by other nations; again, sovereignty issue. Also, what if a nation decides she does not agree to such an agreement?
Everyone intentionally placing their satellites in VLEO is going to be spending more fuel simply staying in orbit against atmospheric drag than they they will spend on avoidance manoeuvres.
This ESA satellite is in VLEO by design, which means every piece of space junk is going to present a threat to it.
Once ESA has secured funding for their navigation automation project (funding to be decided in November), we should hear nothing more about potential collisions with the endless rain of space junk from megaconstellations.
Now it’s true the article used refused, but it does not actually back that up. ESA does not have the authority to tell satellite operateors to do something.
“we informed SpaceX” They where informed of a 1/1,000 risk and declined to take action. Refused implies someone was telling them to do something rather than notifying them of the situation.
The "informed" bit really is telling here. It seems to imply that the author knows ESA didn't "ask" for anything from SpaceX. That or the author doesn't actually know what the communication was.
Either way this is sensationalist and worth a flag.
So? They have that right, what they don't have a right is to demand someone else take action. Just because one party believes something to be necessary does not actually make it so.
If 1, they should be absolutely liable for any damages, considering they actively moved INTO the orbit of the ESA weather satellite in the first place (for their own "testing").
Refuse does imply a certain level of impoliteness or malice. The article only states that spaceX said they wouldn’t be taking any action and later states
> Krag suspected it could be something to do with SpaceX’s electric propulsion system, which “maybe is not reacting so fast” as the chemical propulsion on board Aeolus.
Not acting because you don’t have choice doesn’t count as refusing.
By that premise every space organization should shut down. Nothing should ever go up again. What if something went wrong? What if every mission wasn't perfect? What if someone lost control of a satellite or space station (eg Tiangong-1)? What if people died? (NASA, Soviet Space Program)
They are are where they are by choice. It's a de-orbiting exercise. If they can't maneuver where they are now because of their propulsion design, they have the same problem in a normal orbit. So I wonder whether this design enables them to be a good and safe neighbor.
> the satellites, each weighing 227 kilograms, are “capable of tracking on-orbit debris and autonomously avoiding collision.” But for this incident, this system does not seem to have been used for some reason
"does not seem to have been used" - if SpaceX made no statement we can't really tell? Maybe they were confident that their collision avoidance would simply work.
Yeah, then all SpaceX has to do is run their usual "we have no comment" playbook and the reporter is left with nowhere to go. Convenient for SpaceX I guess!
The reason the author is unaware of the reason is because SpaceX refused to give a reason. Impotence? Incompetence? Malice? Different risk calculation? Game of chicken? Either way if they throw thousands of satellites into space they can't just say "sorry, tragedy of the commons, everyone else should shoulder the risk we created (& the costs to avoid risk)"
SpaceX might not bother giving comment to random bloggers. Especially on Labor day and regarding what they probably regard as a non-incident. There are probably a lot of such individuals asking for comment on any of thousands of different subjects.
You are blaming someone for speculating by speculating about it yourself.
SpaceX is a company with enough money from grants and a grandiose mission “to save the planet” and they can’t afford a PR release at the right time of the week?
Ok? I never said it was wrong to speculate. That's all anyone in this thread is doing, and of course that is fine. As it happened, the real story was just that the message never got to the human Starlink operator. SpaceX didn't refuse, decline, whatever, they just didn't see the email. Just goes to show you that our assumptions about these types of things can be wrong.
Interesting. So if I am the only one to have some information as to why I wanted to move a satellite, you should not publish anything until I tell you why, and so I can simply prevent you from publishing by just not telling you anything. Interesting system.
You could publish that I'm not telling you, but you don't know 'why' I'm not telling you, so that makes you irresponsible too.
SpaceX refused to give a reason. So you can guess: Impotence? Incompetence? Malice? Different risk calculation? Game of chicken?
Either way, if they throw thousands of satellites into space they can't just say "sorry, tragedy of the commons, everyone else should shoulder the risk we created (& the costs to avoid risk)".
Will SpaceX wilfully endanger other satellites (most of which are of higher cost & societal value than an individual SpaceX constellation satellite)? If so, maybe NASA should stop allowing their launches until they offer a better answer.
The issue is constellations like StarLink are setup to tolerate some % of failures and still function. They launch in bulk as part of the overall plan.
So this brings up the question - if the value of your satellite is lower do you burn up precious fuel to manueveur around potential collisions? Or do you play space chicken and let the others maneuver around you?
I think a major topic missed is this 1 in 1000. I think the only chance involved in this, practically speaking, is that the ESAs data about the orbits had enough error in it to not be sure there would be no collision.
This is a lot different than rolling dice or betting on black. Likely there was or was not going to be a collision; run it 1 million times and the results would always be the same. Perhaps Space-X had more optimistic odds.
And yet ESA's risk model said "Move the satellite if the odds are less than 1:10,000". Easy enough for you to say when it isn't your $2.3B satellite, I suppose.
Either possiblity is bad. If the satellite couldn't move, then we are now putting expensive weather satellites at risk because SpaceX has put fixed obstacles into orbit. Furthermore, other organizations have to shoulder fuel costs disproportionately since only they are moving the satellites. SpaceX doesn't have to do anything.
SpaceX has a higher collision tolerance because their satellite is cheap compared to the expensive weather satellite.
Are we really going to allow private companies to bully public institutions?
> Are we really going to allow private companies to bully public institutions?
But on the other hand, should bureaucrats get some sort of celestial "dibs" on certain orbits? In the absence of some other kind of governing body, this is pretty much governed by the law of the jungle. Would you have the same objection if SpaceX had this issue with another company, or are you granting europe some kind of deference because of government?
> But on the other hand, should bureaucrats get some sort of celestial "dibs" on certain orbits?
Yes.
> In the absence of some other kind of governing body, this is pretty much governed by the law of the jungle.
No, we have space treaties.
> Would you have the same objection if SpaceX had this issue with another company, or are you granting europe some kind of deference because of government?
Obviously, that is not in question, it just happens to be ESA this time, it could be NASA then next, or the ISS. Stuff that was already there was not put there with the likes of SpaceX in mind and therefore it is up to the newcomer to accommodate the incumbents. Satellite reaction mass is a severely limited resource.
Yes because things like European weathers satellites serve the public interesting and vital infrastructure needs (GPS, for example). Those have higher priority over a private corporation's satellites.
> If it's the europeans had a lesser tolerance for close passes, why should SpaceX move?
Did you read my first comment? I said that of course SpaceX would have a lower collision tolerance because there satellite is much cheaper/less investment of resources compared to the EU satellite.
Even if SpaceX's satellite was also expensive and SpaceX still had good data, we should prioritize the needs of public institutions that are launching satellites that serve the world over private corporations.
Define public interest. People pay through this through taxes; not really different than paying for spacex whatever via $$.
> less investment
So if SpaceX can do it better, maybe get the bureaucrats and the three-letter alphabet-soup agencies out of the sky, already.
> prioritize the needs of the public institutions
Public institutions have needs and wants. Why is this a need and not a want? This wasn't a military satellite. Why are bureaucrats more important? They accomplish little of significance or merit.
Everytime I see these kind of events I think about Kurzgesagt video [1] about how the domino effect of satellite collisions would release such a high amount space debris that would trap us all on earth forever.
Starlink satellites are only approximately 350 miles up, which means any debris gets brought down by atmospheric drag very quickly. A million satellites in that orbit could not cause kessler syndrome.
That's not quite accurate.
Firstly,the debris will destroy everything in LEO, maybe including ISS, then debit in a few months.
Secondly, some of the debris will probably end up in higher orbit and stay there for a long time, just like was the case with India and China ASAT tests.
Then they can pose a risk for centuries
> Secondly, some of the debris will probably end up in higher orbit and stay there for a long time
The average height of the orbit may go up, but depending on the direction of the collision the perigee will decrease by some amount, so there will still be a considerable amount of atmospheric drag. The Chinese ASAT test was in a much higher orbit, which is why so much of the debris is still troublesome. Most of the debris from the Indian ASAT test is gone, and the rest will probably be gone in a year or two.
It's like when there's a crappy car refusing to give the right a way to a luxury car. The crappy car owner has little to lose compared to the luxury car owner.
Constellation owners (SpaceX here) knows that owners of big satellite will dodge them.
Big satellite will have to carry more fuel, which incidentally is good news for the business of satellite launchers (SpaceX again).
Why would they refuse to move the satellite? It's not like only the European satellite would be affected by a collision, both are in danger, so either they couldn't or they didn't think it was likely.
The likely answer is that the satellite in question was out of control and they couldn't move it.
I don't know which Starlink sat this was, but the the other satellite that did dodge seems to be flying at ~300km. This is much lower than what the functional Starlink constellation is. Some of the initial launch failed to unfold their solar panels, and the batteries would have ran down by now, so they are rapidly decaying space trash.
Others are being deorbited under power but SpaceX would have no reason not to do dodge with those.
I could see them going after a european company, or America going after an American company. I'm guessing the europeans are going to insist they want it done in their courts, and SpaceX is just gonna shrug and say, "Nope, not showing up." europeans can't really make an American company do much.
By international space law full liability is held by the country of launch/license. The US Government would owe the ESA a full paycheck if it was to be damaged. So this is already well covered in international law.
The US military tracking isn't very accurate. Errors of hundreds of meters are common.
I bet spacex themselves know the location of their satellites to within a few meters. With more accurate data, you can come up with a far more accurate collision risk estimate.
Definitely GPS, probably star trackers too. Not a good idea to have a single point of failure on position knowledge when your whole scheme depends on putting small beams in the right places.
My suspicion is that SpaceX and the ESA disagreed either on:
1/ the position of the SpaceX vehicle (ESA may not trust the SpaceX position data)
2/ that they disagreed on the probability of a collision (ESA thinks it's 5%, SpaceX thinks it's 0.005%) or
3/ they disagreed on whether the probability of a collision was worth moving for (eg, ESA thinks you should move for a one-in-a-million shot, SpaceX disagrees).
I recall someone saying that SpaceX did share ephemeris, but I can't find a reference. Maybe I dreamed it.
Being a lover of Space and SpaceX, this incident highlights why a goldilocks paradigm for regulation needs to exist for space. Not too much that it stifles innovation. Not too little that someone ends up Kesslerizing Near-Earth space.
Current regulations are largely successful at promoting commercial space flight, but there is a big transparency gap between civilian space flight agencies and private entities. Transparency should be one of the cornerstones of space regulations, but it isn’t. This incident raises multiple policy questions. Why can’t SpaceX move their satellite? Is it a no because an ion drive takes days to burn or is it a no because they don’t wanna? Either way, it exposes a flaw in SpaceX’s design and raises concerns about StarLink. When there are hundreds of StarLink satellites in the air, will SpaceX say no again if there’s a chance for collision? Will SpaceX be a responsible steward of Near-Earth space? If ion drives are insufficient to address this problem, should they be required to take a small amount of monopropellant as a precaution?
Transparency is essential for answering these questions and asking new ones. The one I’m most concerned about is the question, what are the pollution risks of LEO constellations? Ideally, before StarLink and other networks are up and running, I feel there should be a study of the pollution risk StarLink poses be it light pollution or debris. We shouldn’t repeat the failings of the 20th century in space.
> Either way, it exposes a flaw in SpaceX’s design
What is this statement based on? You are saying yourself that there's no information available to the public about this due to lack of transparency, yet in the next sentence you are making a bold statement about design flaws.
To me it sounds like there should be clear regulation about right of way in space. I haven't seen any information pertaining to SpaceX design so far.
> To me it sounds like there should be clear regulation about right of way in space
Common courtesy worldwide would be for the person causing the risk to move aside first. Worse still it was a private company working on a for-profit product forcing a public weather satellite to divert.
There's no need for regulation until bad actors show up.
Are we then saying there's an equivalent of domain squatting in space? Cause I could see all types of problems showing up if we call that common courtesy.
But the implication is then someone will launch a million tiny satellites doing nothing and occupy a million orbits.
That said, this was in my opinion very bad form for SpaceX, to the point I wonder if, as hinted in the article, they simply can’t do collision avoidance properly.
I don't follow - domain squatting takes a domain and makes it unusable for others, orbit squatting would take an unused orbit and make it unusable for others.
The fundamental question is whether you have the right to the exclusive use of an orbital path just because you were there first. This is unresolved but I'm pretty sure the answer is no.
Assigning value (public/private, research/commercial, etc.) to each parties use just muddies the waters. There needs to be an established resolution process and AFAIK there isn't one.
Domain squatting is taking an unused domain and leaving it unused. Google isn’t domain squatting google.com, just like ESA wasn’t orbit squatting.
Assigning value is also necessary. We reserve spectrum for public safety and eventually may need to reserve orbital paths in the same way. Seems like we are pretty far from filling up the sky even with current avoidance tech though.
No, because satellite orbits aren't owned or held by anyone. Not to mention all low orbits have decent amount of junk that's in the process of deorbiting.
Satellites don't "move" into each other's way. Orbits naturally end up that way. There's no one "causing the risk" in the first place. SpaceX could have just as easily asked the ESA to move their satellite as it was getting in the way of their satellite.
You conveniently avoid the fact that SpaceX inserted itself into a collision course with the ESA’s orbit, long after ESA had been there. Being that ESA has a “right of way” they asked SpaceX to take the action
The risk of a crash was 1 in 1,000. This is 10 times the threshold of 1 in 10,000 that requires a maneuver. Both of these satellites have been in orbit long enough for accumulated error to cause the satellite to drift from its predicted path. There are many satellite orbit-crossing events every day, and the accumulated drift is the primary cause of their orbits becoming close enough to require action. An anticipated collision that requires course correction is as close to a no-fault scenario as there can be.
ESA's orbit is itself constantly changing. Orbits are not fixed. They move constantly from orbital perturbation from Earth's gravity field, the moon's gravity field and, notably at these low altitudes, atmospheric drag. Again, who moved into who's path is not something you're going to determine. Yes SpaceX was lowering its orbit, but it's crossing hundreds of satellites paths by doing that, and even the orbit SpaceX satellites are in are shared by many other satellites. LEO is crowded.
From what I've read so far, it seems that ESA was uncomfortable with the possibility of a collision to a degree that SpaceX could possibly be quite comfortable with - ESA wants > 1km distance, whereas SpaceX might be comfortable with much closer - they are quite different constellations.
Who has the right to dictate who has priority / which criteria is the right one?
I'm going to assume that the cost of ESA's satellite is much greater than SpaceX's StarLink module so obviously they have much more of an interest in preserving full functionality - BUT avoiding collisions is in everybody's best interests. Who is going to use their fuel? Who has right of way, and to what terms?
I think you have this right, not so much about appetite for risk but maybe telemetry on SpaceX's satellite (or a number of satellites in the swarm) gave them sufficient confidence that there was no issue?
Given SpaceX routinely send resupply missions to the ISS I would assume they have a fair amount of confidence when it comes to knowing the position of their vehicles.
... typically of dead satellites or debris that is unable to move itself.
ESA had a weather satellite in orbit, SpaceX actively moved an active satellite _into_ the same orbit as ESA's satellite (to practice maneuvers) and then said "no, we won't move it back out the way".
All deorbiting objects (“satellites” or debris, whatever you want to call them) are going to pass through the orbit of the ESA satellite which is intentionally in a very low orbit.
It is the ESA that actively moved into that orbit by design, intention and active navigation. SpaceX intended their satellite to lift to a higher orbit but it couldn’t make it, so it is being deorbited by atmospheric drag.
All orbits at the same altitude will cross at least twice. This is not a deliberate act of interception but just part of the way orbits work.
That article started that the Spacelink satellite -had- been at a much higher orbit (the "regular" for their satellites, but had been brought down "to test de-orbiting maneuvers". Do you have a source to the contrary?
Also, the ESA satellite has been in that orbit for 8+ years. It wasn't "moved" into the Spacelink orbit by any means.
I get that en route to or from orbit, entities will pass through other orbits.
Deorbiting is exactly the process I was describing.
SpaceX did not deliberately move their craft to interfere with Aeolus. Aeolus is deliberately placed where every deorbiting craft will interfere with it. It’s like someone setting up an office in the foyer of an apartment complex. Everyone entering or leaving the complex is going to walk through the office. They aren’t deliberately walking through the office, it’s just that your office is where everyone is walking to get to where they are going.
In addition, SpaceX is testing deorbiting satellites with Hall effect thrusters that are barely powerful enough to raise orbit from higher in the atmosphere. At this altitude the thrusters would have little or no effect, so the craft is more or less an inert piece of space junk which SpaceX can possibly rotate using reaction wheels and magnetic torquers.
Words have meanings, and there is a vast semantic gulf between “SpaceX refused to take action” and “SpaceX were incapable of taking action”. The author used “refused” but published the article before investigating further. Expect a better story to come from the likes of Loren Grush in a couple of days once people have had time to get a response from SpaceX and digest just what is going on here (NB: ESA budget under review in November, as per ESA’s entire tweet thread of which this social media shitstorm is only responding to the first because it mentions SpaceX)
This is really just a semantic argument so I will leave it here. Reread the article but completely ignore the inflammatory clickbait title and any of the author’s own (emotive, inflammatory) words on the matter, stick to only the quotes from named sources and you should be golden.
> "It is very rare to perform collision avoidance manoeuvres with active satellites. The vast majority of ESA avoidance manoeuvres are the result of dead satellites or fragments from previous collisions."
Basically they think SpaceX is driving recklessly, which is why they put out the press release; they don't do it when they dodge dead satellites.
A collision would result in 1000s of pieces of debris that would affect everyone’s satellites. Repeated collision would make these orbits unusable until gravity and atmospheric drag naturally deorbit the fragments.
Also, even if the impact is limited in term of long lasting debris, losing several satellites, and as a consequence losing the services they provide plus having to replace them is definitely a large impact which will last years.
When it happens we will have to cease all launches for a few years, until the sky clears out. That would certainly be terrible, but not ice-caps-melting-terrible.
The ESA asked SpaceX to move to avoid a collision, and they refused. What would happen if a private airline refused the FAA telling them to redirect to avoid a collision, forcing another aircraft to take evasive action?
It definitely is political, and it's going to become a bigger problem if SpaceX's approach to potential collisions is just to play chicken.
The ESA isnt a governing authority here. The proper analogy would be two planes in uncontrolled airspace on course to come uncomfortably close. One asks the other to change course, and the other refuses. No FAA involved.
This sort of thing actually happens with some regularity. Airliners and other fancy planes carry TCAS systems which can detect other planes and tell the pilot to maneuver to avoid a collision. If the other plane has TCAS, the two units will coordinate to ensure they don’t both dodge the same way. TCAS can also detect planes that don’t have TCAS, which would be virtually all small planes, and in that instance the TCAS-equipped plane will maneuver to avoid on its own. Airliners routinely fly though airspace where not all traffic is under positive control by ATC. Sometimes they end up on a course that sets off TCAS. The airliner will then maneuver to avoid the collision while the light aircraft pilot remains blissfully unaware, unless they happen to see the airliner fly past.
The bigger issue is that ESA is coming up to an important administrative date in November where their future budget will be determined, and they want money for their automation program. So they have to portray space as dangerous, ESA processes as desperately in need of improvement, and this specific example of a super expensive research satellite coming close to part of a future megaconstellation makes for great attention-grabbing headlines.
That it was SpaceX that ESA got to throw under the bus was the cherry on the cake.
I don't see where it positively states that SpaceX provided no reasoning. In my reading of the article, Krag did not speak directly with SpaceX. SpaceX were alerted to the possibility of collision by the U.S. military, and nobody who was party to that conversation is quoted in the article. All we know for sure is that SpaceX did not proactively reach out to the Space Debris Office at the ESA about this issue, at least as of the time the interview with Krag occurred (which would not have been more than a few hours after the incident to get this blog post written and published by 4PM ET given the incident occurred at 7AM ET).
The article explains the ESA is raising this issue because there is no such mechanism to perform “every 5 minutes” avoidance maneuvers right now.
The current process is manual and slow. 22,000 satellites in that orbit mean that picking up the telephone to call each other to say “hey are you going to move your satellite to avoid X possible collision on Y date?” isn’t going to scale, so they’re saying that a better system needs to be put in place.
We’ve seen just recently with multiple US Navy collisions causing loss of life that using “common sense” to avoid collisions instead of being prepared and planning doesn’t always work out well..
The reason why it's a big deal is that SpaceX is planning on putting up a LOT more
On one side is a multimillion-dollar weather satellite, on the other side 1536 throwaway satellites, 5 % of which failed during commissioning (the track record of the initial batch gives you an idea what to expect for the future). It's easy to see who will move their asset out of the way and also easy to see who has no business being up there.
The track record of experimental designs deliberately intended to push the boundaries of reliability really does not give us any idea of what to expect for the future.
Spacecraft engineering is inherently conservative because of reliability concerns, you can't take a tow truck and mechanic in orbit to fix things when it turns out that you didn't understand the latest technology quite as well as you thought.
Space isn't the place to push technical boundaries, also because of orbital littering.
The inability to take a tow truck to orbit is exactly why you need to push the boundaries to find out what works and what doesn’t. Also note that “littering” is an irrelevant analogy in these orbits because of their low altitude. Atmospheric drag will cause anything in these orbits to deorbit in a matter of years to decades.
Just don’t go doing stupid stuff like releasing hundreds of millions of copper needles in a three-thousand-kilometre orbit as part of a misguided attempt to provide a reflective surface to carry military communications around the globe. That would be actual littering, as would any test of an anti-satellite weapon.
The Forbes title frames a bit of bias too. An equally valid title would be "ESA Initially Refused to Move Satellite at Risk of Collission with a SpaceX Satellite".
I don't understand the matter first hand, but I believe this comment I read on reddit today:
The % of fuel used in this type of maneuver is negligible.
A sat in LEO like this uses fuel to:
- correct their orbit from the initial release from the rocket
- maintain orbit
- deorbit
- avoidance
The breakdown is something like: 66%, 33.5%, 0.45%, 0.05%
The ESA sat in this case boosted their sat...
which they planned on doing anyways. So it probably
cost close to nothing. It maybe reduced the lifespan
of the sat by 0.001%
it's a matter of precedent, it's much harder to change rules and agreements when multiple stakeholders have planned on an assumed state of laissez faire, compared to learning from the first "conflicts".
>ESA noted that it performed 28 collision avoidance maneuvers in 2018, but it was mostly to avoid dead satellites or bits of space debris. Maneuvers to avoid active satellites were “very rare”, they said, but the arrival of mega constellations like Starlink raises concerns that many more such maneuvers will be needed in future.
>“What I want is an organized way of doing space traffic. It must be clear when you have such a situation who has to react. And of course automating the system. It cannot be when we have 10,000 satellites in space that there are operators writing the email what to do. This is not how I imagine modern spaceflight.”
SpaceX simply saying it would "not act" is negligent, and I really hope there was more to it that the author just wasn't privy to. If they had better sensors and knew there'd be no collision, say so. If they couldn't maneuver, say so. Otherwise it's just a game of space chicken, risking not only those two sats but all of LEO.
One is a throwaway satellite of a huge constellation - the other a unique billion euro research satellite. Easy to see who will give in first to protect their assets, but plainly it is equally easy to see who should move first.
There's not much to disagree on the odds. Trajectories are known, uncertainties too and therefore probabilities of collisions can be calculated.
Any disagreement here is easy to reconcile, whose assumptions and methods are more accurate?
The disagreement is on the importance of the odds. SpaceX might be OK with a 1:10000 (or whatever) chance of collision, the Europeans want it lower.
The Europeans could in the future force the issue. Move, or we'll fine you and forbid you from launching off of European sites and ban EU companies from using SpaceX.
Considering the cost of launching on Ariane compared to Falcon 9, especially with rideshare upcoming, I don't think EU could ban companies from flying with SpaceX without hurting their own commercial satellite market to an extent.
if there's disagreement on the odds, the entity on the confident side should bet the total value of the other satellite + usage delay costs, in escrow.
Can you describe what action SpaceX should have taken to prevent their inert space debris from potentially colliding with someone else’s active satellite?
Have they stated anywhere they lost control of their satellite, that it was inert? It was under power and reportedly performing a test maneuver.
If they didn’t have control over it they could have easily stated as such to ESA and avoided any bad PR at all, or even just in response to this story.
No need to apologize. I'm not familiar with the Starlink-44's structural design and profile but from promo images that I have seen it would make sense to still consider turning around on xyz axis to reduce impact risk a bit.
You will note that the official ESA comments on the matter are that “nobody did anything wrong,” and (to paraphrase) “we need better technology to remove humans, email and phone calls from the loop”.
For all you know, SpaceX did state to the ESA that their satellite was incapable of performing the required course adjustment. Something along the lines of “our satellite is incapable of adjusting its orbit to accomodate your request.”
As for the author publishing the story before receiving a response from SpaceX, that’s on the author not SpaceX.
> As for the author publishing the story before receiving a response from SpaceX, that’s on the author not SpaceX.
It’s on SpaceX, journalists can’t let companies kill their stories through silence. The story quotes Holger Krag as saying SpaceX replied “they do not plan to take action”. Quite different than “incapable”.
They published the story 4 hours after the incident happened. That's nowhere near enough time for SpaceX to respond, as such, it is indeed on the journalist in my opinion. Not only is it unlikely that the author found out immediately as it happened, it's unlikely that SpaceX could get a statement ready in 2 maybe 3 hours (assuming an hour total between author seeing this info and a SpaceX rep seeing his email about it), while also leaving the author time to write the article.
And yet SpaceX did reply to Loren Grush, who has published a story about this ESA announcement, but it’s quite a different story because she wasn’t setting out to smear SpaceX.
I was under the impression that most of the constellation would be at such an unusually low orbit that there wouldn't be much to collide with. How crowded are the planned orbit shells?
What's interesting is that there's already a similar internationally agreed framework for how to approach this - shipping - https://en.wikipedia.org/wiki/International_Regulations_for_... - These would probably be a good place to start considering a framework for satellite orbit adjustments. However, despite this framework, it still happens - and I can see the issue of risk evaluation being a big one - what if one entity thinks a 1/1000 risk is acceptable for their cheap satellite vs the other satellite wanting 1/1000000?
You're liable for damage to the other party if you're at fault for a maritime collision, so there's at least some countervailing economic pressure against taking the risk. Your insurers will not be happy with you if you negligently sink someone else's oil tanker.
I agree with this. Rules of the seas are not laws but more like a code for mariners to follow.
Too bad we see such ignorance in space.
Think of the ESA bird as a "boat in anchor". Lit up with the proper anchorage lights.
Now some jackass with a jetski is blasting right twords you.
> ESA is preparing to automate this process using #AI
Sounds like something SpaceX and other constellation satellite companies should do as part of these widescale deployment. Or at least lower the threshold of moving out of the way so they don't get tons of push back from existing space power players who’ve got more influence than they do.
This is gonna get ugly pretty soon. I'm pretty convinced it's less an issue of moving (which esa itself stated it does once every week to two weeks) and more of politics. A couple possibilities:
* europe mad at America b/c trade
* europe mad at American companies beating european ones to space and getting more of the limited orbit spots first
* europe mad she wasn't really asked before SpaceX put up satellites
* europe has a principal objection to the idea of private companies playing on the same level as the bureaucrats or an issue with corps occupying space at all
Most likely, it's a combination of all of the above. Regardless, people will start getting awfully catty about this, and it's possible knocking some one else's satellite out of the sky could be a serious issue.
NORAD 44278 is a Starlink satellite presently being de-orbited by SpaceX in a controlled fashion to simulate end-of-life disposal. [1]
The current altitude (~330km) [2] puts it well below the operation envelope and it’s currently the lowest sat of the Starlink constellation at this point.
It appears SpaceX does continue to have some level of control of the craft. [3] Some speculation I’ve read is that the much higher impulse of the ESA thrusters would make the ESA craft a better candidate to execute the maneuver.
These adjustments are not rare - they happen approximately every other week for this particular craft, and as Iridium says, they do this weekly. [4]
A video of the conjecture [5] shows an approach of “less than 10km”. The ESA craft apparently passed much closer to several other craft over the last week, so there may be more to it in this case, or there may be a bone to pick with Starlink.
Lastly, ESA has a particular interest in raising this issue now as much as possible, as these “mega-constellations” launched by US companies are coming closer to reality and the current processes (picking up the phone) don’t scale well to tens of thousands of satellites.
The “more to it” is that ESA is gearing up to request funding for their AI project, billed as an automated tracking and avoidance manoeuvre planning system capable of handling the vast numbers of objects and debris expected to be in orbit in the near future.
This was not about “SpaceX refused to move” but “we need money to build this automated system so we don’t have to do it all by hand,” and an additional level which is, how should the various satellite operators make their orbital parameters available to interested parties?
The situation is that ESA didn’t even know what kind of object the track represented, and they had to contact SpaceX to ask about it in the first place. Their job would have been easier (whether manual or automated) if SpaceX made the orbital parameters available for automated queries.
The last thing this is about is SpaceX being “arrogant” or “refusing to move a satellite.”
> The last thing this is about is SpaceX being “arrogant” or “refusing to move a satellite.”
SpaceX entering ESA’s orbit and arrogantly refusing to move is an excellent showcase of the need for better avoidance systems, especially since they plan to launch 12k of them.
It’s also an excellent warning of SpaceX’s ethical framework.
If the Starlink satellite is moving (I don't mean actively moving, just that it's changed its orbit in the last few days), the measurements of its orbit are going to be less precise. So even if expected closest distance is greater, the error bars are wider. Which would give a larger probability of a conjunction.
I guess when you have your own launch capabilities you can treat your internet satellites as easily replaced.
Adding more junk up there only serves to create more launches as the inevitable collisions occur. It arguably makes good business sense for SpaceX to add as much as possible and be quite fast and loose with their management of it, at least as long as it's the wild west up there.
The long term solution would be to establish some form of property rights over particular orbits. Then, to use a particular orbit, one must pay the market rate for it.
Certainly property rights to orbits will have special characteristics. We've already accumulated much experience with property rights for things that are different from physical property - I'm sure spending more than 30 seconds thinking about it can come up with solutions.
This is a false article. There is no evidence SpaceX refused to move and ESA calling out the fact that they moved for a SpaceX satellite is ESA playing politics. LEO is full of space debris and satellite owners move satellites for space junk all the time. This is called political grandstanding.
228 comments
[ 2.9 ms ] story [ 253 ms ] threadEDIT: Said comment is now one full screen tall due to manual insertion of linewraps, and is very wasteful of space on desktop and landscape layouts: https://i.imgur.com/H15eCXo.png
As the number of satellites in orbit increases, due to 'mega constellations' such as #Starlink comprising hundreds or even thousands of satellites, today's 'manual' collision avoidance process will become impossible...
I find it foreboding that the number of intersections with space debris and live constellations is to the point where ESA is moving towards non-deterministic algorithms to hopefully avoid the worst. How will Musk ever leave this planet if we make a shell of debris around ourselves?
See also: the anime "planetes"
Since all these orbital parameters change over time due to atmospheric drag, solar pressure, gravity irregularities, etc, organizations like NORAD constantly monitor and determine the orbital parameters of every object they can find circling the Earth.
So, they watch for when the likelihood that one satellite will come within some small distance (say, 5km) of another is high. For example, if the odds that they come that close are say, 75% in the next 5 orbits, the operators are notified and alter their courses slightly to maintain separation.
Even manned spacecraft such as the ISS will perform separation maneuvers to keep themselves clear of objects that may come too close.
And it's slightly discomfiting to realize that just traffic control is so complex that it needs modern computers to track and react. Space is supposed to be vast and empty! And it is, but not in LEO.
SHARCs with frackin’ laser beams
I'm actually wondering if there's not a need to communicate more between all the bodies having satellites in space - just like autonomous cars, if both make an evasive maneuver they might cancel each other out. So there needs to be some kind of communication to avoid this.
I think that's a phenomenon much older than autonomous car theory.
https://www.urbandictionary.com/define.php?term=Hallway%20da...
- there are also uncontrolled debris, non-tracked objects, military satellites with unpublished orbits etc
- just like with automating road vehicle traffic, it would require very high degree of centralization, which is never going to happen due to various reasons
- the propellant is limited so you also want to reduce the amount of avoidance maneuvers to a minimum
So no, statistical models (which you probably mean by AI) won't improve the situation much. What matters is:
- increasing the orbit determination accuracy and coverage with better instruments and commercial tracking networks; NORAD TLEs alone are not precise enough for reliable collision prediction
- better space weather prediction
- improving the coordination between the operators
- deployable decelerators for dead satellites in lower orbits, also possibly debris removal if the problem becomes really serious
i.e. stuff most operators already do, except for the debris removal
> Calculations made by CelesTrak had expected these two satellites to miss by 584 meters (1,916 ft)
Perhaps the calculations aren't straightforward Newtonian mechanics?
* https://en.wikipedia.org/wiki/2009_satellite_collision
I wasn’t suggesting to not use AI algorithm for computing the collision probabilities. For this, they will obv use whatever tools they already have.
The whole discussion is around how to react automatically to a likely collision.
Regarding that particular collision, IIRC human factor was involved a lot. Iridium has been warned well in advance, but hesitated to make the decision since their models gave lower probabilities.
I'm as big a fan of SpaceX as you'll find but considering they were the one entering an orbit they didn't previously occupy it seems like a pretty lame move on SpaceX's part.
It puts them in a bad light, and that could impact their ability to get necessary licenses to operate.
Sounds like a fantastic barrier to entry for newcomers who don't have billions lining their pockets.
This ESA satellite is in VLEO by design, which means every piece of space junk is going to present a threat to it.
Once ESA has secured funding for their navigation automation project (funding to be decided in November), we should hear nothing more about potential collisions with the endless rain of space junk from megaconstellations.
So, they refused. That's what that word means.
“we informed SpaceX” They where informed of a 1/1,000 risk and declined to take action. Refused implies someone was telling them to do something rather than notifying them of the situation.
Either way this is sensationalist and worth a flag.
Well, somebody seems to have needed to maneuver.
> they ... declined for reasons
"Declined" and "refused" are synonyms. SpaceX did refuse to move the satellite.
> Krag suspected it could be something to do with SpaceX’s electric propulsion system, which “maybe is not reacting so fast” as the chemical propulsion on board Aeolus.
Not acting because you don’t have choice doesn’t count as refusing.
It does beg the question whether you should be there at all.
By that premise every space organization should shut down. Nothing should ever go up again. What if something went wrong? What if every mission wasn't perfect? What if someone lost control of a satellite or space station (eg Tiangong-1)? What if people died? (NASA, Soviet Space Program)
> the satellites, each weighing 227 kilograms, are “capable of tracking on-orbit debris and autonomously avoiding collision.” But for this incident, this system does not seem to have been used for some reason
"does not seem to have been used" - if SpaceX made no statement we can't really tell? Maybe they were confident that their collision avoidance would simply work.
> Maybe they were confident that their collision avoidance would simply work.
Just like Tesla's? If your satellite is worth $100K and the other one is worth $2.3B, perhaps you have a little less to lose.
This is such an incomplete report, it reminds me of the old adage "A lie can get halfway around the world before the truth can put it's pants on."
A proper, less sensational reporter would have waited until they knew why they declined to move the satellite before writing the piece.
SpaceX is a company with enough money from grants and a grandiose mission “to save the planet” and they can’t afford a PR release at the right time of the week?
You could publish that I'm not telling you, but you don't know 'why' I'm not telling you, so that makes you irresponsible too.
Either way, if they throw thousands of satellites into space they can't just say "sorry, tragedy of the commons, everyone else should shoulder the risk we created (& the costs to avoid risk)".
Will SpaceX wilfully endanger other satellites (most of which are of higher cost & societal value than an individual SpaceX constellation satellite)? If so, maybe NASA should stop allowing their launches until they offer a better answer.
So this brings up the question - if the value of your satellite is lower do you burn up precious fuel to manueveur around potential collisions? Or do you play space chicken and let the others maneuver around you?
With their electrically propelled Tesla.
This is a lot different than rolling dice or betting on black. Likely there was or was not going to be a collision; run it 1 million times and the results would always be the same. Perhaps Space-X had more optimistic odds.
SpaceX has a higher collision tolerance because their satellite is cheap compared to the expensive weather satellite.
Are we really going to allow private companies to bully public institutions?
But on the other hand, should bureaucrats get some sort of celestial "dibs" on certain orbits? In the absence of some other kind of governing body, this is pretty much governed by the law of the jungle. Would you have the same objection if SpaceX had this issue with another company, or are you granting europe some kind of deference because of government?
Yes.
> In the absence of some other kind of governing body, this is pretty much governed by the law of the jungle.
No, we have space treaties.
> Would you have the same objection if SpaceX had this issue with another company, or are you granting europe some kind of deference because of government?
Obviously, that is not in question, it just happens to be ESA this time, it could be NASA then next, or the ISS. Stuff that was already there was not put there with the likes of SpaceX in mind and therefore it is up to the newcomer to accommodate the incumbents. Satellite reaction mass is a severely limited resource.
Why?
> space treaties
Which cover some public entities and no private ones.
> there first
But SpaceX was in a different orbit, and felt fine. If it's the europeans had a lesser tolerance for close passes, why should SpaceX move?
> If it's the europeans had a lesser tolerance for close passes, why should SpaceX move?
Did you read my first comment? I said that of course SpaceX would have a lower collision tolerance because there satellite is much cheaper/less investment of resources compared to the EU satellite.
Even if SpaceX's satellite was also expensive and SpaceX still had good data, we should prioritize the needs of public institutions that are launching satellites that serve the world over private corporations.
Define public interest. People pay through this through taxes; not really different than paying for spacex whatever via $$.
> less investment
So if SpaceX can do it better, maybe get the bureaucrats and the three-letter alphabet-soup agencies out of the sky, already.
> prioritize the needs of the public institutions
Public institutions have needs and wants. Why is this a need and not a want? This wasn't a military satellite. Why are bureaucrats more important? They accomplish little of significance or merit.
[1] https://www.youtube.com/watch?v=yS1ibDImAYU
Secondly, some of the debris will probably end up in higher orbit and stay there for a long time, just like was the case with India and China ASAT tests. Then they can pose a risk for centuries
The average height of the orbit may go up, but depending on the direction of the collision the perigee will decrease by some amount, so there will still be a considerable amount of atmospheric drag. The Chinese ASAT test was in a much higher orbit, which is why so much of the debris is still troublesome. Most of the debris from the Indian ASAT test is gone, and the rest will probably be gone in a year or two.
Constellation owners (SpaceX here) knows that owners of big satellite will dodge them. Big satellite will have to carry more fuel, which incidentally is good news for the business of satellite launchers (SpaceX again).
I don't know which Starlink sat this was, but the the other satellite that did dodge seems to be flying at ~300km. This is much lower than what the functional Starlink constellation is. Some of the initial launch failed to unfold their solar panels, and the batteries would have ran down by now, so they are rapidly decaying space trash.
Others are being deorbited under power but SpaceX would have no reason not to do dodge with those.
There's very few large American companies for whom the EU is neither a huge market, nor a major location.
I bet spacex themselves know the location of their satellites to within a few meters. With more accurate data, you can come up with a far more accurate collision risk estimate.
Controlling the target makes some problems a lot easier.
Do they have really good star trackers on their sats?
[1]: https://space.stackexchange.com/questions/563/in-leo-what-is...
Once they have a good orbit fit, they will propagate it to predict the probability of collision a few days in advance.
Details towards the end: https://techcrunch.com/2019/05/24/spacex-reveals-more-starli...
GPS works extremely well in low earth orbit.
Also these aren't military assets, their specs and locations aren't secret, so if they have better tracking, they should be sharing that.
Do we know that for sure?
1/ the position of the SpaceX vehicle (ESA may not trust the SpaceX position data)
2/ that they disagreed on the probability of a collision (ESA thinks it's 5%, SpaceX thinks it's 0.005%) or
3/ they disagreed on whether the probability of a collision was worth moving for (eg, ESA thinks you should move for a one-in-a-million shot, SpaceX disagrees).
I recall someone saying that SpaceX did share ephemeris, but I can't find a reference. Maybe I dreamed it.
Current regulations are largely successful at promoting commercial space flight, but there is a big transparency gap between civilian space flight agencies and private entities. Transparency should be one of the cornerstones of space regulations, but it isn’t. This incident raises multiple policy questions. Why can’t SpaceX move their satellite? Is it a no because an ion drive takes days to burn or is it a no because they don’t wanna? Either way, it exposes a flaw in SpaceX’s design and raises concerns about StarLink. When there are hundreds of StarLink satellites in the air, will SpaceX say no again if there’s a chance for collision? Will SpaceX be a responsible steward of Near-Earth space? If ion drives are insufficient to address this problem, should they be required to take a small amount of monopropellant as a precaution?
Transparency is essential for answering these questions and asking new ones. The one I’m most concerned about is the question, what are the pollution risks of LEO constellations? Ideally, before StarLink and other networks are up and running, I feel there should be a study of the pollution risk StarLink poses be it light pollution or debris. We shouldn’t repeat the failings of the 20th century in space.
What is this statement based on? You are saying yourself that there's no information available to the public about this due to lack of transparency, yet in the next sentence you are making a bold statement about design flaws.
To me it sounds like there should be clear regulation about right of way in space. I haven't seen any information pertaining to SpaceX design so far.
Common courtesy worldwide would be for the person causing the risk to move aside first. Worse still it was a private company working on a for-profit product forcing a public weather satellite to divert.
There's no need for regulation until bad actors show up.
That said, this was in my opinion very bad form for SpaceX, to the point I wonder if, as hinted in the article, they simply can’t do collision avoidance properly.
The fundamental question is whether you have the right to the exclusive use of an orbital path just because you were there first. This is unresolved but I'm pretty sure the answer is no.
Assigning value (public/private, research/commercial, etc.) to each parties use just muddies the waters. There needs to be an established resolution process and AFAIK there isn't one.
Assigning value is also necessary. We reserve spectrum for public safety and eventually may need to reserve orbital paths in the same way. Seems like we are pretty far from filling up the sky even with current avoidance tech though.
What a ridiculous statement. I suppose you'd also suggest no need for traffic laws until there's a 50 car pileup?
Bad actors are sometimes involuntarily so.
Don’t we know what needs doing? Why do we have to reinvent the wheel?
From what I've read so far, it seems that ESA was uncomfortable with the possibility of a collision to a degree that SpaceX could possibly be quite comfortable with - ESA wants > 1km distance, whereas SpaceX might be comfortable with much closer - they are quite different constellations.
Who has the right to dictate who has priority / which criteria is the right one?
I'm going to assume that the cost of ESA's satellite is much greater than SpaceX's StarLink module so obviously they have much more of an interest in preserving full functionality - BUT avoiding collisions is in everybody's best interests. Who is going to use their fuel? Who has right of way, and to what terms?
Given SpaceX routinely send resupply missions to the ISS I would assume they have a fair amount of confidence when it comes to knowing the position of their vehicles.
Typical 'better to ask permission than forgiveness' fuckery that companies featured on this site are famous for.
https://twitter.com/IridiumBoss/status/1168582141128650753
> Hmmm. We move our satellites on average once a week and don't put out a press release to say who we maneuvered around... -Matt Desch, CEO of Iridium
ESA had a weather satellite in orbit, SpaceX actively moved an active satellite _into_ the same orbit as ESA's satellite (to practice maneuvers) and then said "no, we won't move it back out the way".
All deorbiting objects (“satellites” or debris, whatever you want to call them) are going to pass through the orbit of the ESA satellite which is intentionally in a very low orbit.
It is the ESA that actively moved into that orbit by design, intention and active navigation. SpaceX intended their satellite to lift to a higher orbit but it couldn’t make it, so it is being deorbited by atmospheric drag.
All orbits at the same altitude will cross at least twice. This is not a deliberate act of interception but just part of the way orbits work.
Also, the ESA satellite has been in that orbit for 8+ years. It wasn't "moved" into the Spacelink orbit by any means.
I get that en route to or from orbit, entities will pass through other orbits.
SpaceX did not deliberately move their craft to interfere with Aeolus. Aeolus is deliberately placed where every deorbiting craft will interfere with it. It’s like someone setting up an office in the foyer of an apartment complex. Everyone entering or leaving the complex is going to walk through the office. They aren’t deliberately walking through the office, it’s just that your office is where everyone is walking to get to where they are going.
In addition, SpaceX is testing deorbiting satellites with Hall effect thrusters that are barely powerful enough to raise orbit from higher in the atmosphere. At this altitude the thrusters would have little or no effect, so the craft is more or less an inert piece of space junk which SpaceX can possibly rotate using reaction wheels and magnetic torquers.
Words have meanings, and there is a vast semantic gulf between “SpaceX refused to take action” and “SpaceX were incapable of taking action”. The author used “refused” but published the article before investigating further. Expect a better story to come from the likes of Loren Grush in a couple of days once people have had time to get a response from SpaceX and digest just what is going on here (NB: ESA budget under review in November, as per ESA’s entire tweet thread of which this social media shitstorm is only responding to the first because it mentions SpaceX)
This is really just a semantic argument so I will leave it here. Reread the article but completely ignore the inflammatory clickbait title and any of the author’s own (emotive, inflammatory) words on the matter, stick to only the quotes from named sources and you should be golden.
I don't understand this. If the satellite is being de-orbited, why bother pissing off the ESA to save fuel?
Basically they think SpaceX is driving recklessly, which is why they put out the press release; they don't do it when they dodge dead satellites.
They've also refused to comment, and Musk's reputation doesn't engender faith in their reasons.
Also, even if the impact is limited in term of long lasting debris, losing several satellites, and as a consequence losing the services they provide plus having to replace them is definitely a large impact which will last years.
So this approach won't work.
It definitely is political, and it's going to become a bigger problem if SpaceX's approach to potential collisions is just to play chicken.
This sort of thing actually happens with some regularity. Airliners and other fancy planes carry TCAS systems which can detect other planes and tell the pilot to maneuver to avoid a collision. If the other plane has TCAS, the two units will coordinate to ensure they don’t both dodge the same way. TCAS can also detect planes that don’t have TCAS, which would be virtually all small planes, and in that instance the TCAS-equipped plane will maneuver to avoid on its own. Airliners routinely fly though airspace where not all traffic is under positive control by ATC. Sometimes they end up on a course that sets off TCAS. The airliner will then maneuver to avoid the collision while the light aircraft pilot remains blissfully unaware, unless they happen to see the airliner fly past.
You're not wrong, but annoying the folks that may be involved in regulatory approval for your venture to get commercially licensed is not prudent.
That it was SpaceX that ESA got to throw under the bus was the cherry on the cake.
The SpaceX satellite is for all intents and purposes inert, since it is incapable of significantly adjusting its orbit.
You know the old joke about the US Navy warship giving way to the second class seaman, right?
If SpaceX is actively deorbiting the satellite, it's not inert.
"Blue, we are uncomfortable with the possibility of collision on our current courses, requesting you adjust course 5 degrees starboard."
"Nah Red, looking all good here."
"Okay, we will move." - Every 5 minutes on the water probably
more like "nah." spacex did not inform ESA that the risk is lower or anything like that.
The current process is manual and slow. 22,000 satellites in that orbit mean that picking up the telephone to call each other to say “hey are you going to move your satellite to avoid X possible collision on Y date?” isn’t going to scale, so they’re saying that a better system needs to be put in place.
We’ve seen just recently with multiple US Navy collisions causing loss of life that using “common sense” to avoid collisions instead of being prepared and planning doesn’t always work out well..
The reason why it's a big deal is that SpaceX is planning on putting up a LOT more.
On one side is a multimillion-dollar weather satellite, on the other side 1536 throwaway satellites, 5 % of which failed during commissioning (the track record of the initial batch gives you an idea what to expect for the future). It's easy to see who will move their asset out of the way and also easy to see who has no business being up there.
Space isn't the place to push technical boundaries, also because of orbital littering.
Fortunately space, for over half a century, has routinely been the place to push all manner of technical boundaries.
Just don’t go doing stupid stuff like releasing hundreds of millions of copper needles in a three-thousand-kilometre orbit as part of a misguided attempt to provide a reflective surface to carry military communications around the globe. That would be actual littering, as would any test of an anti-satellite weapon.
>Normal part of flying satellites (or should be). Does BA announce every time one of its airplanes steers around a Ryanair aircraft?
>“What I want is an organized way of doing space traffic. It must be clear when you have such a situation who has to react. And of course automating the system. It cannot be when we have 10,000 satellites in space that there are operators writing the email what to do. This is not how I imagine modern spaceflight.”
SpaceX simply saying it would "not act" is negligent, and I really hope there was more to it that the author just wasn't privy to. If they had better sensors and knew there'd be no collision, say so. If they couldn't maneuver, say so. Otherwise it's just a game of space chicken, risking not only those two sats but all of LEO.
Any disagreement here is easy to reconcile, whose assumptions and methods are more accurate?
The disagreement is on the importance of the odds. SpaceX might be OK with a 1:10000 (or whatever) chance of collision, the Europeans want it lower.
The Europeans could in the future force the issue. Move, or we'll fine you and forbid you from launching off of European sites and ban EU companies from using SpaceX.
safe? let Musk put his money where his mouth is.
If they didn’t have control over it they could have easily stated as such to ESA and avoided any bad PR at all, or even just in response to this story.
For all you know, SpaceX did state to the ESA that their satellite was incapable of performing the required course adjustment. Something along the lines of “our satellite is incapable of adjusting its orbit to accomodate your request.”
As for the author publishing the story before receiving a response from SpaceX, that’s on the author not SpaceX.
It’s on SpaceX, journalists can’t let companies kill their stories through silence. The story quotes Holger Krag as saying SpaceX replied “they do not plan to take action”. Quite different than “incapable”.
https://www.theverge.com/2019/9/3/20847243/spacex-starlink-s...
Sounds like something SpaceX and other constellation satellite companies should do as part of these widescale deployment. Or at least lower the threshold of moving out of the way so they don't get tons of push back from existing space power players who’ve got more influence than they do.
* europe mad at America b/c trade
* europe mad at American companies beating european ones to space and getting more of the limited orbit spots first
* europe mad she wasn't really asked before SpaceX put up satellites
* europe has a principal objection to the idea of private companies playing on the same level as the bureaucrats or an issue with corps occupying space at all
Most likely, it's a combination of all of the above. Regardless, people will start getting awfully catty about this, and it's possible knocking some one else's satellite out of the sky could be a serious issue.
The current altitude (~330km) [2] puts it well below the operation envelope and it’s currently the lowest sat of the Starlink constellation at this point.
It appears SpaceX does continue to have some level of control of the craft. [3] Some speculation I’ve read is that the much higher impulse of the ESA thrusters would make the ESA craft a better candidate to execute the maneuver.
These adjustments are not rare - they happen approximately every other week for this particular craft, and as Iridium says, they do this weekly. [4]
A video of the conjecture [5] shows an approach of “less than 10km”. The ESA craft apparently passed much closer to several other craft over the last week, so there may be more to it in this case, or there may be a bone to pick with Starlink.
Lastly, ESA has a particular interest in raising this issue now as much as possible, as these “mega-constellations” launched by US companies are coming closer to reality and the current processes (picking up the phone) don’t scale well to tens of thousands of satellites.
[1] - https://twitter.com/thesheetztweetz/status/11446893347775242...
[2] - https://www.n2yo.com/satellite/?s=44278
[3] - https://twitter.com/planet4589/status/1168637324592328705?s=...
[4] - https://twitter.com/iridiumboss/status/1168582141128650753?s...
[5] - https://twitter.com/m_r_thomp/status/1168583891726344193?s=2...
This was not about “SpaceX refused to move” but “we need money to build this automated system so we don’t have to do it all by hand,” and an additional level which is, how should the various satellite operators make their orbital parameters available to interested parties?
The situation is that ESA didn’t even know what kind of object the track represented, and they had to contact SpaceX to ask about it in the first place. Their job would have been easier (whether manual or automated) if SpaceX made the orbital parameters available for automated queries.
The last thing this is about is SpaceX being “arrogant” or “refusing to move a satellite.”
SpaceX entering ESA’s orbit and arrogantly refusing to move is an excellent showcase of the need for better avoidance systems, especially since they plan to launch 12k of them.
It’s also an excellent warning of SpaceX’s ethical framework.
Clickbait headlines are poison, as are hearsay and anonymous sources using emotive language like “SpaceX refused” to move their satellite.
Adding more junk up there only serves to create more launches as the inevitable collisions occur. It arguably makes good business sense for SpaceX to add as much as possible and be quite fast and loose with their management of it, at least as long as it's the wild west up there.