Slashdot Mirror
Satellite Loaded With AI For Self-Diagnosis
TheReckoning writes "NASA has loaded its E0-1 Satellite with Artificial Intelligence to diagnose on-board failures. The software 'works by comparing a computerized model of how the spacecraft's systems and software should perform against actual performance. If the spacecraft's behavior differs from the model, then the ... "reasoner" looks for the root cause of this difference and gives flight controllers several suggestions of what might have gone wrong.' Another NASA probe loaded with AI was Deep Space 1."
Is this AI at all?
Reminds me of Mostly Harmless by Douglas Adams. The robots retrieve the backup computer core for the spaceship to fix the cracked one, and fall through an unknown hole into space carrying it, as an asteroid hit the ship, leaving a big hole and a cracked computer core.
Note, that probably could have been said better, but nothing can do proper justice to Douglas Adams but himself.
Hurricane Ivan: A 17th century prison collapsed. All of the inmates escaped.
Just wait until this thing collides with that other space probe that is programmed to eliminate organic entities off the outside of spaceships, and they combine their AI into a super-probe that is out to exterminate all life.
Ooh! Scary!
fifth sigma, inc.
It just seems to me that it would be better to install more sensors, data-gathering, and reporting capabilities and then leave the trouble-shooting to the people on the ground. Payload costs are expensive, so why put the diagnostic end in orbit?
Given the same data and placed groundside, it could then it could be tuned and upgraded more easily.
I recall how the Mars lander had problems and the ground team worked out a novel solution. I'll bet that they would have like to had extra information to work with, instead of an onboard AI.
BTW - I can understand this approach better for a long-range craft, just not an orbital satellite.
Sure, any automated response can be called AI, but this doesn't impress me.
If all an AI module can do is make objective suggestions, it's nothing more than a list of conditional statements. Whoopideedoo!
I can run similar "AI" on my TI-85. And I could write it all from scratch in the time it takes for a launch vehicle to reach the stratosphere.
The web servers of 10 years ago could "suggest" that an "Object may have moved", so is that artificial intelligence? I guess it's really, really dumb AI....
There have been a few comments about how this should be run on the ground, and it's a waste to be put on a satellite.
Well, similar systems run on the ground now; we know how to do that. Why, in science, should we be content to continue doing things we know how to do? That may sound like a joke, but we need to do this in order to progress.
This stuff, and future versions, will be essential for long range human missions, but it has to be tested now. The bugs should be worked out by running it close by.
Then when it's used for a long range mission, with humans on board, it is less likely to kill everyone because it hadn't been put through its paces.
4096R/EF7BAFA6 79E1 DF98 D09D 898F 9A11 F6F0 DDDC 23FA EF7B AFA6
How the hell does an AI malfunction? Is it a mechanical hardware? That's like asking, what if an algorithm malfunctions. Somethings don't just malfunction, it may contain a logic error, the hardware may malfunction, but software doesn't malfunction, it always does what it is told to do, the way it is told to do so.
------ Curiosity killed the cat. {satisfaction brought it back | it didn't die ignorant | lack of it is killing mankind
I'm reminded of some of the fly-by-wire computer control systems used in aircraft. From what I understand, multiple computer are used, each designed to carry out the same task, but each designed in isolation by a different team of engineers.
Since they are designed in isolation from each other, they are not exactly the same in terms of the way they are built(even though they carry out the same tasks) and thus if one computer malfunctions and gives a response different to the others, the other computers "out vote" this malfuntioning unit.
you have just hit on one aspect of Godel's Incompleteness theorum. The more complex you make a system in a qwest to be perfect, the more paradox that you create. In this case, you have an infinite number if AIs, all of which can fail, and since you can never know the state of the end of the chain, you will never know if the error can be successfull resolved.
HA! I just wasted some of your bandwidth with a frivolous sig!
what happens if the AI malfunctions? then mission control will get a bunch of useless error reports...
[sarcasm] Yep, I'm positive that the hundreds if not thousands of PhD-level man hours that went into this part of the project didn't consider that. Yep, took that young whippersnapper Quasar1999 to think about it for a few mintues to evaluate and assess the entire effort and proclaim, "it's a stupid self diagnosis test." [/sarcasm]
If one actually reads the referenced article, it sounds like LV2 is, in fact, something far more advanced than a "stupid self diagnosis test." Se.f-diagnosis tests are pretty straightforward and highly tuned to a specific architecture. I've written something like that to evaluate an experimental compiler, with statements like,
define a=1;
if (a+a eq 2) then print 'simple addition works'
But LV2 is very differnt than that. Into LV2 (which, despite the hype in the article, does not need to be on-board) is built a generic model of satelite functionality customized to the particular device in question. When unexpected results are found, the diagnostic software can experiment on the model, asking questions like, "if, in the model, valve G34 is stuck open, does the model behavior match the current anomalous condition?" I'm sure it wouldn't be hard to write up a test script that could iteratively simulate a fault in one or more parts of the system until it found a handful of likely candidates. Given that there are thousands of components in a satelite, this surely can be done faster by a machine than by a human. Then, were we really trying to do something advanced, we might come up with a way of caching these results to guide future diagnoses and build up a set of experiences. Collect these experiences from different projects (since, if LV2 and its descendent software is widely adopted, the data are presumably in common form), and you can guide designs of future satelites to avoid common failure modes, or identify problematic components.
Now, is that AI? Does it think? You probably wouldn't say so. Could it be an aid to ground-based support? You betcha. Is there a reason to disrespect the fine engineers at NASA by demeaning their efforts without giving fair due? I fail to see one.
Put my fist through my alarm clock with its ding-dong death inside my ear. - The Blackjacks.
In space, software may malfunction because of a cosmic ray hitting a CPU or bit in memory in just the wrong location and flipping a bit. This is why there are usually three flight computers in a spacecraft: to detect and recover from these transient errors.
"Artificial Intelligence" is simply a fancy way of stating "I don't understand it." To us, a machine like Deep Blue is an example of AI, but to the designers, it's simply a basic state machine. Once you know the output generated from every input, the machine in question is no smarter than your average coffeemaker. Granted, some machines can make coffee faster than others, but that doesn't mean they are smarter.
The same rules apply to humans. The only reason we can consider ourselves intelligent is because we don't understand ourselves. The fact that we do this is a major reason why we can't understand ourselves, which makes us intelligent.