Slashdot Mirror
SpaceX's Latest Advantage? Blowing Up Its Own Rocket, Automatically (qz.com)
SpaceX has reportedly worked with the Air Force to develop a GPS-equipped on-board computer, called the "Automatic Flight Safety System," that will safely and automatically detonate a Falcon 9 rocket in the sky if the launch threatens to go awry. Previously, an Air Force range-safety officer was required to be in place, ready to transmit a signal to detonate the rocket. Quartz reports: No other U.S. rocket has this capability yet, and it could open up new advantages for SpaceX: The U.S. Air Force is considering launches to polar orbits from Cape Canaveral, but the flight path is only viable if the rockets don't need to be tracked for range-safety reasons. That means SpaceX is the only company that could take advantage of the new corridor to space. Rockets at the Cape normally launch satellites eastward over the Atlantic into orbits roughly parallel to the equator. Launches from Florida into orbits traveling from pole to pole generally sent rockets too close to populated areas for the Air Force's liking. The new rules allow them to thread a safe path southward, past Miami and over Cuba.
SpaceX pushed for the new automated system for several reasons. One was efficacy: The on-board computer can react more quickly than human beings relying on radar data and radio transmissions to signal across miles of airspace, which gives the rocket more time to correct its course before blowing up in the event of an error. As important, the automated system means the company doesn't need to pay for the full use of the Air Force radar installations on launch day, which means SpaceX doesn't need to pay for some 160 U.S. Air Force staff to be on duty for their launches, saving the company and its customers money. Most impressively, the automated system will make it possible for SpaceX to fly multiple boosters at once in a single launch.
SpaceX pushed for the new automated system for several reasons. One was efficacy: The on-board computer can react more quickly than human beings relying on radar data and radio transmissions to signal across miles of airspace, which gives the rocket more time to correct its course before blowing up in the event of an error. As important, the automated system means the company doesn't need to pay for the full use of the Air Force radar installations on launch day, which means SpaceX doesn't need to pay for some 160 U.S. Air Force staff to be on duty for their launches, saving the company and its customers money. Most impressively, the automated system will make it possible for SpaceX to fly multiple boosters at once in a single launch.
âoeItâ(TM)s not a bug itâ(TM)s a feature. âoe
Well played SpaceX. Well played.
So they are going to try and close Vandenberg AFB and take a chunk out of California's economy?
I'm sure it's been sorted but this comes to mind:
Reports Say U.S. Drone was Hijacked by Iran Through GPS Spoofing.
(The nabbing of a drone by spoofed GPS signals)
In order to spoof GPS for a rocket you'd have to have a system that had multiple nodes at various altitudes along the exact flight path in order to have a strong enough signal to overpower the real satellites... it seems extremely unlikely that something going as fast as a rocket could be spoofed, unlike a drone which is usually sent to basically hover over an area.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
BOOM ;)
I quote for Slashdot posterity a long and informative piece of relevant information from many years ago, because I fear it's disappearing from the web:
Reliability of Shuttle Destruct System [LONG]
"MARTIN J. MOORE" [mooremj@eglin-vax]
28 Jan 86 14:06:00 CDT
Copyright © 1986 Martin J. Moore
[COMMENT: READERS -- PLEASE OBSERVE THE RESTRICTIONS ON THIS MESSAGE AT THE END OF THE MESSAGE. PGN]
> From: Peter G. Neumann [Neumann@SRI-CSL.ARPA]
> For those of you who haven't heard, the Challenger blew up this morning...
> One unvoiced concern from the RISKS point of view is the presence on each
> shuttle of a semi-automatic self-destruct mechanism. Hopefully that
> mechanism cannot be accidentally triggered.
[COMMENT: I did not intend to imply that as the cause -- only to raise concern about the safety of such mechanisms. PGN]
Peter, I assume that you are talking about the Range Safety Command Destruct System, which is used to destroy errant missiles launched from Cape Canaveral. From 1980 to 1983 I was the lead programmer/analyst on the ground portions of that system, and I am the primary author of the software which translates the closing of destruct switches into the RF destruct signals sent to the vehicle. I think I can address the question of whether the system can be accidentally triggered; worrying about that gave me nightmares off and on for months while I was on the project. I'd like to tell you a little about the system and why I think the answer is No. Note that my information is now three years old, and some details may have changed; there may also be minor errors in detail due to lapses in my memory, which isn't as good as my computer's!
On board the vehicle, there are five destruct receivers: one on the external tank (ET) and two on each of the solid rocket boosters (SRBs). There is no receiver or destruct ordnance on the Orbiter; it is effectively just an airplane. The casing of each SRB is mined with HMX, a high explosive; the ET contains a small pyrotechnic device which causes its load of liquid hydrogen and liquid oxygen to combine and combust. The receivers and explosives are connected such that the receipt of four proper ARM sequences followed by a proper FIRE sequence by any of the receivers will explode the ordnance.
The ARM sequence and FIRE sequence must come from the ground; they cannot be generated aboard the vehicle. These sequences are transmitted on a frequency which is reserved, at all times, for this purpose and this purpose alone. There are several transmitters around the Eastern Test Range which can be used to transmit the codes. These transmitters have a power of 10 kw (continuous wave). The ARM and FIRE sequences consist of thirteen tone pairs (different for each command and changed for each launch). There are eight possible tones, resulting in 28 possible tone pairs; thus, there are (28^13) or slightly over 6.5E18 correct sequences.
The Range Safety Officer has two switches labeled "ARM" and "DESTRUCT". When he throws a switch, it generates an interrupt in the central processor (there are actually two central processors running and receiving all inputs, but only one is on-line at any time; in case of software or hardware error the backup is switched in. And yes, they have different power sources.) The central program checks for the correct code on each of two different hardware lines (the correct code is different for each line); if correct, and all criteria are met to allow the sequence to be sent, the central program requests the tone pairs for that sequence from another processor. That processor (like everything else in the system, actually redundant processors) has only one function: to store and deliver those tone pairs. The processor resides in a special vault and can only be accessed in order to program the tone pairs (which are highly classified) before each launch. The data line between the central processor and the storage processor is
Just a little reminder that there are plenty of missiles out there that are NOT nukes. And a lot of them aren't nearly as destructive as a Falcon Heavy going boom right over your house. Here's hoping they have a good flight.
Don't disappoint your bird dog. Go to the range.
You only like it because you don't work in the reliability field. Having a human operator in charge is one of the least reliable ways of doing things.
I swear I heard ~15 years ago that (at least some) NASA rockets utilized a gyroscope to automatically detonate during launch if they started pointing below the horizon.
Corruption is convincing someone that the selfless ideal is the same as their selfish ideal.
So instead of a person bring responsible for pressing a button there is now AI making that decision.
Better get used to that. It's our future whether we like it or not.
NASA and the Air Force (which provides the range safety systems) have been working on the autonomous flight safety system for at least a decade. SpaceX is just the first customer to use it.
...that all rocket explosions are automatic. They're rarely intended or desired but they still qualify as "automatic."
=Smidge=
I was under the impression that Ariane 5 did automatically self-destruct in 1996.
"Launches from Florida into orbits traveling from pole to pole generally sent rockets too close to populated areas for the Air Force's liking. The new rules allow them to thread a safe path southward, past Miami and over Cuba."
Actually, according to my research, Miami and Cuba are in fact populated areas.
i could live a little longer in this prison
It's not an AI, it's an algorithm. And likely not even one using machine learning.
If construction was anything like programming, an incorrectly fitted lock would bring down the entire building...
I don't know. GPS was never supposed to be used for anything like this.
*Civilian* GPS was not supposed to be used like this and got limitations (speed, altitude *) to avoid being usable like this.
The military had guiding missile in this way in their mind from day one.
---
*: normal GPS chips will refuse to give a precise answer above a certain speed (~500 m/s) and altitude (18km).
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
Prior to iOS 11, you had to hold down the quote button to get the option to use "smart quotes". Now that those are the default, holding the button down may give the option to use standard quotes. If not, one can turn them off entirely in Settings > General > Keyboards.
Which one is closer depends very much on how long after launch we're talking about. It's a space rocket - toward the end of the flight is very much nearer space than it is to the ground. In fact the Falcon may go twice as high as GPS satellites.
It's OK, those Latinos are communists, and anyone who would say communist lives matter must be a dirty commie!
"When information is power, privacy is freedom" - Jah-Wren Ryel
No it's more than that. With the old system, someone on the ground had to monitor the flight path of the rocket and press the self-destruct if it went far enough off course. This meant you needed to be able to track the entire flight of the rocket with radar ground stations (which they can't for this kind of launch), and you had to hope the self-destruct signal from the ground got through to the rocket.
This new system eliminates both problems, because the rocket tracks itself (using onboard GPS sensors and a predefined safe flight path) and autonomously triggers its own self-destruct if it exits the safe flight path.
If SpaceX can get permission to use encrypted military GPS signals (an off-the-shelf civilian GPS is programmed not to output data beyond 1200mph or 60,000ft anyway) they can also be safe from any GPS spoofing attempts. Otherwise a terrorist group could launch some high-altitude balloons above the flight path carrying GPS spoofing devices to cause a false positive leading to detonation, possibly raining debris on inhabited areas. Or it could be a way for one of the few eco-terrorist groups out there to make a statement.
"When information is power, privacy is freedom" - Jah-Wren Ryel
The only manned flight these days is to the Space Station which is inclined about 51 degrees; that's apparently "roughly parallel to the equator" according to the BS summary. Polar orbits are used for weather and spy satellites.
More evidence of AI stealing jobs.
W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
Am I the only one that read it "AI Force" instead of "Air Force"?
You only like it because you don't work in the reliability field. Having a human operator in charge is one of the least reliable ways of doing things.
Then again, there are the famous Airbus incidents where software caused the plane to safely mow through a forest and crash because it knew that the pilot desperately trying to fly it was obviously wrong.
Or the computer glitch that told another Airbus that it was somehow flying nose-up at 30 degrees at cruising speed, and immediately pitched it down at 30 because it then thought it was in level flight. Miraculously they eventually wrested control and managed to land - though safely is a bit strong of a word for that mess.
There was another case where software was added to keep pilots from throttling back right after takeoff. Some did this to lower noise. There was an airstrike on the flight as the plane took off. The engines were running rough, the pilot called for reduced power, but the software insisted on full power for takeoff. Killed the engines and the plane had to land in a local field.
The point is, I don't know that I'd take the position that the human is the least reliable ways of doing things, when the humans tried to do the correct thing, but the computers insisted on their way or the highway.
I'd like to know how they determined that this is failsafe. The presumably unsafe range officers - do you have the numbers of flights that should have been destroyed that were not? The only one I know of was a Chinese launch that took out a nearby town, and I don't even know if that had a human in the loop or was a more reliable computer
TL;DR - Don't be in too big a hurry to declare superior safety. Hubris always attracts Karma
The shepherds did so well protecting the flock that the sheep no longer believed that wolves existed.
No, we call it an emergency self-destruct system. A rocket is already a missile by nature, with its fuel being the warhead. If it were to malfunction and hit the ground with most of its fuel still on board it would make for a *really* bad day for anyone in the area. A high altitude airburst as soon as the situation becomes unrecoverable is by far the preferable alternative.
--- Most topics have many sides worth arguing, allow me to take one opposite you.
Then again, there are the famous Airbus incidents where software caused the plane to safely mow through a forest and crash because it knew that the pilot desperately trying to fly it was obviously wrong.
On indeed. No computer is perfect, and no system created by people is perfect. In industry we look at the differences between random failure and systematic failures. Some >80% of failures of systems are systematic and the result of human error in design, operation or maintenance. The remainder can be easily quantified and is widely considered several orders of magnitude better in performance than humans.
The point is, I don't know that I'd take the position that the human is the least reliable ways of doing things, when the humans tried to do the correct thing, but the computers insisted on their way or the highway.
I'm reminded of the usual safety pep talks: No one goes to work with the intent to injure themselves (obviously not true, but true enough). If you consider humans doing the correct thing then they are actually quite reliable. However the key reliability problem is that humans startlingly often don't do the correct thing, often due to no fault of their own. The human brain is incredibly fallible.
TL;DR - Don't be in too big a hurry to declare superior safety. Hubris always attracts Karma
Safety systems were invented for a reason and humans are only ever considered the first line of defence before automatic systems take over. I often like getting asked why I don't perform reliability calculations on emergency stop pushbuttons on critical equipment. The answer typically stops the person asking the question dead in their tracks: "Without doing a calculation I can say the reliability of the pushbutton is approximately 3 orders of magnitude higher than the brain that is tasked with making the decision to push it."
Space is only 60 miles away. GPS satellites are ~12,000 miles away. You'll be in stable orbit long before you get closer to the satellites. At which point automated self destruct systems will almost certainly be disengaged because there's no longer an imminent threat to anyone, have essentially limitless time to try to regain control, and any explosion is going to create some nasty orbital debris that nobody wants around.
--- Most topics have many sides worth arguing, allow me to take one opposite you.
Safety systems were invented for a reason and humans are only ever considered the first line of defence before automatic systems take over. I often like getting asked why I don't perform reliability calculations on emergency stop pushbuttons on critical equipment. The answer typically stops the person asking the question dead in their tracks: "Without doing a calculation I can say the reliability of the pushbutton is approximately 3 orders of magnitude higher than the brain that is tasked with making the decision to push it."
I think the takeaway here - or at least the one that to me would work best, is to have both. An override of the computer in the event it refuses to destroy an obviously errant rocket. Which should actually increase safety, not just eliminate payroll.
The shepherds did so well protecting the flock that the sheep no longer believed that wolves existed.
I'm Canadian. I keep reading it as "Air Farce".
#DeleteFacebook
Polar orbits are used for weather and "weather" satellites.
FTFY.
Signed,
CIA/FBI/NSA/X-Files/whatever.
#DeleteFacebook
That's not a quite good analogy.
Here we are talking about replacing a proven (albeit expensive) method with a cheap computer on board of the rocket.
Those computers have always been there in the past, the range safety explosives (and any eventual rescue systems in the case of manned flights - e.g. the Apollo/Soyuz escape towers) could have been either triggered automatically when the electronics detected an anomalous deviation from the pre-programmed path OR remotely by a human from the ground should something have gone so wrong that the computer didn't (or couldn't - e.g. the Challenger explosion where the errant boosters were destroyed from the ground after the fuel tank explosion destroyed the orbiter) handle it.
The system was designed with redundancy in mind. Now we are cutting one side of that redundancy to save costs. If you work in safety systems, you certainly know that there are plenty of examples from the past where such simplification and cost savings have costed lives - e.g. the infamous Therac 25 (hw interlocks replaced with software - several people dead from radiation overdoses and many more injured), original DC 10 cargo door design where the elements indicating whether the door was safe or not were changed to show only whether it was "supposed" to be safe and not an actual position of the interlocking (door blew out at altitude, causing an airplane crash that killed 300+ people), plenty of railway accidents have been caused by a loss of redundancy in the signalling systems over the years, etc.
It doesn't take that much imagination to envision a situation where e.g. an unfortunate lightning strike knocks the computers irrecoverably offline (freak accidents that shouldn't happen but they do - e.g Apollo 12 where they got uber lucky). And now you can only watch as those 500 tons of steel and explosive fuel are going to uncontrollably land somewhere, with no way to do anything about it.
Could the current method be made less expensive? Most likely, the airforce (or military in general) are rarely concerned with doing things in a cost effective way. But taking the human out of the loop completely? That doesn't sound like such a smart idea. His role is not to be reliable (that's the computer's job) but to guard against the unexpected.
It takes about two seconds to realize that any ground based system will be passed WAY out of range in about a second at the speed a rocket is going.
Not to mention the signal from a ground based station would be whack because of how fast the rocket is moving relative to the station, a GPS satellite being pretty far away means a rocket can lock on and track it very quickly even going fairly fast; no way the GPS circuits would be able to lock onto the rapidity receding ground station that is my comparison super close, it simply would never factor in.
But apparently you are too ignorant as to how GPS systems these days work... don't feel too bad, many of the people on the Earth are probably almost as stupid as you. Maybe.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
Uhh, no, GPS was designed specifically to guide missiles and military aircraft.
I'm a good cook. I'm a fantastic eater. - Steven Brust
At 60 miles, the air pressure is very low. That doesn't mean you have "limitless time" or any of that. In order to orbit at that altitude, you'd need to be traveling at 20KM/ s or so. The Falcon is only going 500 m/s at that altitude. It would need to be going about 40 times as fast for what you said to make sense.
Challenger's boosters are an excellent case for automated range safety.
How long did they spin out of control for? Video suggests 15 to 20 seconds. Or about a hundred times longer than an automated range safety device would have let them, greatly increasing the debris field.
You are in a twisty maze of processor lines, all alike.
There is a lot of hype here.
These are great examples of times when the computer messed up the day.
You can't make an informed decision based on that alone (well, you can, but science can't).
For balance, we now need to list every time the computer saved the day and see which method really wins.
the External Tank contains a small pyrotechnic device which causes its load of liquid hydrogen and liquid oxygen to combine and combust.
I have often wondered -- and this makes me wonder once again -- how did the Challenger's H2 and O2 become an explosive mixture? Inside the External Tank were actually a LOX tank (above) and a separate LH2 tank (below).
The cause of the disaster was explained as a faulty O-ring that allowed a jet of hot gas to escape out the side of one of the solid rocket boosters, impinging on the External Tank. Fine, but that could cause a breach of the O2 tank or the H2 tank -- not both.
And even if both were breached, wouldn't the breaches have to happen in a very unlikely configuration, to allow the escaping O2 and H2 to travel toward each other and become an explosive mixture?
Maybe the Destruct System played some part in this after all.
That that is is that that that that is not is not.
Yes and no. Ideal redundancy is through independent systems and by independent I am also talking technologically independent. For basic and defined outcomes a human is not needed. A human is critical in decision making where the decision point isn't well defined (i.e. does that car look like it's about to run a red light).
The payroll in this case is a bunch of humans looking at a computer to make a decision based on an event. If the decision point is defined then the computer can just cut out the middle man (no pun intended).
Here we are talking about replacing a proven (albeit expensive) method with a cheap computer on board of the rocket.
The cost or proven in use case have no bearing on the overall reliability of a system. Computer based safety systems have by their nature replaced proven and expensive methods of safety that came before them, yet with each iteration of technology the reliability continues to improve.
What it sounds like you're making a case for is the lack of field experience for this particular computer, but that's exactly where reliability engineering comes in, something that we have also gotten far better at over the years.
Any event that happened to the vehicle that knocked the computers offline would have caused the same impact to the flight termination system regardless of if it was manual or automated. The ability to manually terminate the flight still exists, they have just now added the ability for the vehicle to decide to trigger the FTS itself. If there is any reduction in potential safety, it would come from switching from radar tracking to GPS tracking, not the vehicle having the ability to push the button itself.
These are great examples of times when the computer messed up the day. You can't make an informed decision based on that alone (well, you can, but science can't). For balance, we now need to list every time the computer saved the day and see which method really wins.
Well, I suppose if you are trying to say that I'm wanting to go back to woodburning rockets.
My point is that people have a tendency to believe that a human in the loop equals bad, and that computers will always be accurate.
I'm dealing with that very thiing right now as we write. There is a process that currently involves hand checking a number of databases.
This is being replaced with an automated checking process that is demonstrably less accurate than the hand checked version.
I can demonstrate this easily, but there is a powerful meme at work that says even if hand checking is more accurate, it is less accurate.
But that's how people are - the biggest advantage is that the hand checks are challenged all the time, around 20 percent. but the acceptance rate on the automated version is 100 percent. So whatever, do it and move on to the next project. Fortunately, this is not the sort of thing that causes People to die, but if they are going to launch these things fairly close to populated areas when doing polar orbit launches, they better hope that the always more accurate computer controlled goboom works every time.
Then again, a Rocket that lands in say Jacksonville because the computer won't kill it - I'm certain that everyone will be completely understanding, yes? It was just an outlier. https://www.youtube.com/watch?...
The shepherds did so well protecting the flock that the sheep no longer believed that wolves existed.
[quote] how did the Challenger's H2 and O2 become an explosive mixture? [A jet of hot gas] could cause a breach of the O2 tank or the H2 tank -- not both.[/quote]
The mixture was caused by kinectic force. When the H2 tank breached the rupture rocketed it into the O2 tank.
It has been a while since I studied them but I think the ITAR regulations only apply if the GPS receiver is exported as with cryptography.
According to the Wikipedia segment I've linked, it's indeed an import/export rule.
So in theory an pure 100% all-USAmerican chip manufacturer (do such thing still exist ?) can legally flash a non limited firmware as long as the device never cross US' border during production, is only sold in-land, and is clearly market "not for export".
Also means that the usual asian chip manufacturer only need to flash such firmware on thing clearly sold elsewhere but not in the US (nor the few other countries which follow these rules). They could still flash non-limited firmware for other market as long as they mark them "not for export to US and XyZ countries" (i.e.: only sell them on the less obvious corners of Ali Baba)
If this is an issue for SpaceX, then they should have no trouble getting a licensed exception and I assume some of the ASIC manufacturers produce a custom firmware without the civilian ITAR restrictions
Yup, very likely that SpaceX will easily get the proper licensing, given their field of work. (It's a completely legit use, and the AirForce is on this with them anyway).
Still, they're probably the first US civlian company to be able to basically get a GPS on a giant Missile-like device.
(Until then, US civilian use has been restricted by the export regulation).
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]