Mars rover is frozen in place following software error | Hacker News Reader

Mars rover is frozen in place following software error

extremetech.com

148 pointsjonbaer70 comments

NASA is now reporting the problem as fixed: https://mars.nasa.gov/msl/mission-updates/8589/sols-2653-265...

(Press article at https://www.space.com/mars-rover-curiosity-attitude-glitch.h...)

A better title would be: Mars rover stops moving because it lost track of its orientation. This is uncommon, but well within the realm of "normal" contingencies. And in fact it has already been fixed.

: The Curiosity Story

Having worked on autonomous robot software, I'm willing to bet this is software taking the blame for a hardware issue :-)

Having worked on autonomous robot hardware i beg to differ

Definitely a wetware problem.

Problem exists between keyboard and chair. Or soldering iron and bench.

The hardware and software both sit between the soldering iron and bench.

Proves my point, right? :)

Ah yes, the programmers blame the engineers, and the engineers blame the programmers

Let's not forget the compilers. They're always messing stuff up!

You don't blame the tool. You blame the toolmaker.

Every real toolmaker is gifted with perfect precognitive capabilities so their tool behaves exactly as it should for all users, don'cha'know.

Naw. It's always user error.

I should have included a /s, but was feeling a bit cheeky.

I'm sure you've had to deal with plenty of spurious bug reports in your career :-)

When Einstein worked on nuclear bomb the saying was: If he fails French will say he's American, Americans will say he's German, and Germans will say he's Jew. If he succeed Germans will say he's German, Americans will say he's American, and French will say he belongs to the world.

His comment was about the theory of relativity rather than the bomb:

https://en.wikiquote.org/wiki/Albert_Einstein#1910s

Einstein didn't work on the nuclear bomb at all. According to https://www.amnh.org/exhibitions/einstein/peace-and-war/the-..., he wrote that famous letter to Roosevelt and was later denied the necessary security clearance.

Amateurs. Real programmers blame other programmers.

The productive programmers I've met, and they are pretty up there, tend to blame it on the documented specs.

Jesus who wrote this crap code?

Looks at commit history

Oh it was me...

Past-me is such an idiot...

We should both know better

I think the hardware engineers ought to be blaming the physicist, keep the blame going up the abstraction ladder. Then the physicists can blame the universe, which was created in a lab by software engineers.

Well clearly sir you are wrong and haveth no idea of which you speak

Having worked on neither, I blame it on someone trying to impress a lady.

Big Bang Theory reference? When Howard was trying to impress a girl?

https://bigbangtheory.fandom.com/wiki/Mars_Rover

yep.

Having worked with both autonomous robot hardware and software engineers. I’m certain it’s a communication problem. Not between software and hardware mind you, one between humans.

No burn, but having worked outside the domain of autonomous robots, I'm willing to place the blame on an autonomous robot issue

As a hardware engineer it doesn't matter if it is a hardware bug. Software should work around it.

I have been working designing both hardware and firmware for autonomous robots for the last 6 years.

When asked if a problem is hardware or software you provide the same answer if someone asks you if you are a god... you say YES!

A developer I know at JPL has been quite stressed about this operation. Unfortunately I can’t speak to details of the software changes.

It's had quite a few NAND-related issues over its time.

That is the nature of the job, because the payoff when things work is so rewarding.

Good luck, Curiosity developers, engineers, and scientists -- we are all rooting for you.

Would appreciate any technical info from someone who knows more about this, the article is not targeted to a technical audience.

Here is a description of the problem from a UC Davis geobiologist who is using the rover to study habitability.

https://mars.nasa.gov/msl/mission-updates/8587/sols-2649-265...

> Thanks to Curiosity, we have a better idea of where water existed (and may still exist) on Mars, as well as where we might be able to find evidence of ancient life.

I don't understand why everyone is so excited about finding life on Mars.

Life on Mars means life may be relatively common on water-bearing planets, which means the Filter is more likely to ahead of us than behind.

https://www.youtube.com/watch?v=sNhhvQGsMEc

Life on Mars could be related to life on Earth as both bodies have traded mass quite regularly over the last few billion years.

We don't have any evidence of life being common.

True. But we have prima facie evidence that advanced civilizations -- as advanced as we're about to be, in not very much longer -- are very rare.

Finding life on other planets means life was probably seeded on earth instead of the 'Spontanous Generation' theory of biology where RNA just happened on earth at some point. It also opens up the possibilities of life existing outside of earth. I don't know whats not mind blowing about that.

As a biologist, it's pretty cool, although not quite as cool because it's the same life. Different life, evolved on different planets, would be cooler. It would use some of the same tools, like phosphates as signaling/energy molecules, because nothing else can do the job. But alien life could use different amino acids from the ones we use, or something different from amino acids instead. They would probably be linear polymers of related monomers though. The same goes for DNA: different base units probably, but still a linear polymer, and very possibly a helix (lots of things are helices).

But the problem with all these instances of life is that all but n of them have to get snuffed out before they've developed much further than we are, and so far as we know n has to be either 1 or 0.

Don’t pressures and temperatures impact which chemical classes can do which job?

It depends on how far out you want to get.

It's possible life could exist on neutron stars, or in stellar photospheres, or in the atmospheres of gas giants. And life might not be obvious. Life as we know it is not an obvious prediction given carbon, hydrogen, oxygen and nitrogen as a starting point. So maybe it turns out that certain combinations of minerals do something interesting when you immerse them in an ocean of liquid chlorine. We wouldn't know.

But so far as we know for sure, life has to start with liquid water, which narrows the range of admissible temperatures to 0-100 C and pressures around 1 atmosphere. Hyperthermophile bacteria can live in boiling water (up to 120C). We use some of their enzymes to do genetic engineering on normal cells. The temperature change isn't enough stop the hyperthermophile proteins from functioning (and they're pretty sensitive). Likewise there are psychrotolerant (cold-dwelling) cells that can live near 0C, using mostly normal proteins.

So I think life as we know it is locked into one regime of "normal" chemical behavior.

Also as a biologist,finding life on other planets..no matter what the form...would be mind blowing as fuck to me. Sorry you're so jaded.

Do we have evidence of that? It seems like we've not looked in sufficient detail at a sufficient number of solar systems to reach a conclusion of "very rare".

I agree, absence of evidence is not evidence of absence.

Life anywhere outside our planet would definitely be a big deal. Mars makes a good candidate to at least find traces of past life there

there's no reason the Great Filter couldn't be both ahead and behind us.

- Life on planets with fossil fuels available advances to an industrial phase and dies off either from nuclear war or global warming

- Life on planets without fossil fuels never advances to even the iron age.

You're right that fossil fuels are contingent. They exist because of a big evolutionary coincidence that might not be repeated on other worlds.

It's possible, though, that aliens might not need fossil fuels to get to space. If they had more efficient photosynthesis, for example (ours is <5% efficient), living plants could have the energy content of gasoline. They might eat gasoline. All their fuel usage would be carbon neutral (CO2 pulled out of the atmosphere). They would be way ahead of us.

Or it's possible that we're the most advanced because fossil fuels are necessary and nobody else got them.

How efficient would these plants need to be?

Wait, can you elaborate on that? Cursory glance at google suggests there can be sufficient other sources of hydrocarbons aside from dinosaurs being wiped out by a meteor.

Plants evolved lignin well before fungus figured out how to break it down. That gap in time is when most of our coal was formed.

Afaik the iron age depended on charcoal, not on fossil fuels. Europe has seen a wide deforestation due to it. Industrialization might depend on fossil fuels but not an iron age per se.

The idea of the 'great filter' isn't that a necessary item's absence prevents advancement -- the idea is that through technological advancement, something happens to destroy most civilizations along the way.

but stated it as you have, there's a huge number of filters and requirements, and they are all over the place, and that's definitely true.

It will be very exciting to learn how life on Mars is different or similar to life on Earth. It would teach us what are might be fundamental blocks of life, and what is optional or can be "implemented" differently. It might bring us closer to understand beginnings of life.

Even if we find that life on Mars has common roots with life on Earth, it will be gives us new perspective on early life forms plus we will learn that lifeforms could survive interplanetary trips without special protections.

If the life we find on Mars isn't DNA-based, but something entirely different, it would suggest that DNA-based life may be unique to Earth, and that every planet hosting life will likely be fundamentally unique.

That's pretty much the only reason I get excited about finding life on other planets.

Sort of. Life on Mars would provide an instance of life AND filter, since that life has not progressed over billions of years and is likely extinct. And that filter would be very likely behind us, given life on mars was very likely more primitive than it is on Earth.

Or it could be after Mars and before Cambrian.

This seems to be no big deal. Stuff like this is how learn to do real world robotics.

Sure, but your heart is probably still going to skip a few beats when you realize a software glitch has halted a $2.5 billion device.

Absolutely. But these types of glitches are necessary to learn about real world problems.

And then to resume to slightly-above normal heart beats when you remember it was designed for way less time/usage than it actually survives. Guess we'll see if this is the final stone or not, apparently they are trying to get it right again (thinking about that, wonder how it feels to deploy a hotfix to another fucking planet)

Viewed another way, a successful bugfix brings a $2.5B instrument back to life.

And it demonstrates that the design allowed bug fixes while the rover is on another planet. I view this as big success for the engineering. They did a great job. It’s more important to have the ability to fix bugs than avoiding bugs.

Assuming your bug fix mechanism is bug free...

True. That's where a lot of the quality effort should go. Also keep this part as simple as possible.

...on another planet

Can’t we just send a few humans already? The wouldn’t be able to stay there for years, but it’d be a few orders of magnitude more efficient.

What a shitty title. The entire article doesn't even contain the word software.

Reply on news.ycombinator.com