Why the Perseverance Rover computer is so "outdated"

[8tg]

Comparing a consumer computer to the computer on a Mars rover is like comparing the fuel economy of a Toyota Prius and an M1 Abrams.

 

Okay sure, your Prius gets 57.4mpg more than my Abrams, but is your Prius’s use case to decimate entire countries? Or to go to the grocery store?

[Blue4130]

1 hour ago, Grand Admiral Thrawn said:

@Fabioo It might also be interesting for you that ISS crew uses 20+ year old Thinkpads and DSL speed onboard connection.

Iss has 600mb/s high-speed. Double what I have.

[harryk]

NASA Technology Readiness Level Guidelines

NASA uses TRL ratings to qualify hardware for missions. For a high value mission like Perseverance NASA won't accept anything less than TRL-9 (maybe TRL-8) for any critical hardware, e.g. primary computer. Most COTS hardware, even stuff designed for spaceflight, is only TRL-4 until picked up for a mission. It often takes years just to complete the validation to reach flight readiness not including the time needed to redesign when the hardware does not pass.

 

Interestingly though, the Ingenuity helicopter, being purely experimental, carries significantly more computing power than the rover.

https://spectrum.ieee.org/automaton/aerospace/robotic-exploration/nasa-designed-perseverance-helicopter-rover-fly-autonomously-mars

Quote

Tim Canham: Since Ingenuity is classified as a technology demo, JPL is willing to accept more risk. The main unmanned projects like rovers and deep space explorers are what’s called Class B missions, in which there are many people working on ruggedized hardware and software over many years. With a technology demo, JPL is willing to try new ways of doing things. So we essentially went out and used a lot of off-the-shelf consumer hardware. 

 

There are some avionics components that are very tough and radiation resistant, but much of the technology is commercial grade. The processor board that we used, for instance, is a Snapdragon 801, which is manufactured by Qualcomm. It’s essentially a cell phone class processor, and the board is very small. But ironically, because it’s relatively modern technology, it’s vastly more powerful than the processors that are flying on the rover. We actually have a couple of orders of magnitude more computing power than the rover does, because we need it. Our guidance loops are running at 500 Hz in order to maintain control in the atmosphere that we're flying in. And on top of that, we’re capturing images and analyzing features and tracking them from frame to frame at 30 Hz, and so there's some pretty serious computing power needed for that. And none of the avionics that NASA is currently flying are anywhere near powerful enough. In some cases we literally ordered parts from SparkFun [Electronics]. Our philosophy was, “this is commercial hardware, but we’ll test it, and if it works well, we’ll use it.”

 

[Heliian]

I work with avionics, you would be shocked at how low tech most production items are.  Windows NT ain't dead, it's just up here with me. 

 

But yes, reliability is key.  A lot of people ask why we just don't use iPads to run everything, they're reliable right?  Yes, in a non life and death way. When you entrust big missions and/or peoples lives with it, it better have almost zero failure rate and multiple redundancies on board and a never ending self test loop or two or three.

 

 

[Morrie Sells Wigs]

4 hours ago, ki8aras said:

pardon me?

Pardon me, what?

 

Which part of my comment did you have trouble with?

[Morrie Sells Wigs]

3 hours ago, FloRolf said:

Don't feed the trolls m8. 

If I had a different opinion to yours would that mean I'm trolling?

[PlayStation 2]

I'm not into space tech - but I think having something that has as many kinks worked out of it as possible is preferable to something that's bleeding edge when it comes to something being sent to a whole 'nother planet.

[ki8aras]

2 minutes ago, Maury Sells Wigs said:

Pardon me, what?

 

Which part of my comment did you have trouble with?

all that we're saying is that we don't see how your opinion has any base on reality

[LloydLynx]

• PowerPC 750

My Power Mac has been the most stable computer I've ever used, so that checks out for reliability.

[HerrKaLeu]

All the calcs use relatively little computing power. I mean they flew to the Moon in the 1960's with a computer that has less power than many watches nowadays. Watch the movie " hidden figures", they had computers to prepare the mission. By computers I mean people whose job it was to compute with pen and paper. I'm sure that 200MHz system the rover has can run circles around all the human and mainframe computers they had. 

 

And you don't find the newest technology AND proven track record in the same device. That thing also flies through millions of miles of space radiation. Mars also has less sun-power for the solar cells (more distance from sun). So power consumption is even more of an issue. Design of that rover probably started 10-15 years ago. They can't be like certain manufacturers and use the consumer as the beta tester

 

Same can probably said about the OS and software it uses. It will look very primitive compared to Windows, but it will work ALL THE TIME. No one from NASA wants to come to this forum mid-mission and ask advice on how to re-boot...

 

[LloydLynx]

1 minute ago, SGT-AMD said:

Can it pass the 2g test? Vibration test?

Temp test?

Consumer grade is not NASA grade

 

I'm only talking about processing stability. I've never had to deal with an unresponsive window, Firefox just kept chugging along to finish what I wanted of it and was fully responsive the whole time, try browsing the modern web on a Pentium 3 and it'll constantly be going unresponsive. Something about that architecture is just so dang solid.

[Zodiark1593]

7 hours ago, Fabioo said:

I get your points, but I'm not sure if you are right on that one. All the landing phases had to be done remotely, with no earth inout whatsoever. All the sensors onboard the landing aircraft used machine learning (from the rover and from the landing crane) so plenty of processing happened on the rover.

Doesn't shielding mitigate that?

It’s quite possible that pre-calculated tables were utilized to reduce processing load. There’s plenty of RAM to do something of the sort. 
 

Additionally, the use of a radiation-hardened CPU (RAD750) as the main brains doesn’t preclude the employment of faster DSP or VFP units to offload heavier tasks when necessary as well. While it isn’t yet terribly feasible to have very fast processors that are thoroughly radiation-hardened, shielding and redundancy can allow the use of faster components to take on more demanding workloads when the time comes. 
 

That said, I don’t think you’d need much more than Playstation 2 levels of performance (which is about what RAD hard processes seem to be about). It wouldn’t be surprising if there were some radiation hardened off board DSP units already available. 
 

I’d definitely like to see someone with the CPU engineering prowess of Apple (for example) design a space-ready SoC. Though validation and testing will take many years, it would be intriguing to see what engineers can do with current radiation-hardened  fabrication. 

[Morrie Sells Wigs]

2 hours ago, ki8aras said:

all that we're saying is that we don't see how your opinion has any base on reality

All I'm saying is I don't see how your opinion has any basis in reality. 

[Mihle]

6 hours ago, porina said:

On a similar note someone had asked why better cameras aren't used in space. The short answer was that the limit is in the data path of getting that data back to Earth. They only have a certain amount so generating much more than that doesn't help. Also there was a case of, they have a camera and know it works.

Curiosity had 2 MP camera, but the Perseverance actually have a 20 MP camera. Cameras are a thing they have updated.

 

I am guessing the CPU is the exact same one as in Curiosity.

[RejZoR]

It's same as with consoles or iPhones. People just look at bigger numbers of everything and think bigger is better. Just like consoles or iPhones, stuff for such missions is very purpose built. It's not like this rover will have bunch of stupid tasks running in background and with Google eating most of RAM and bandwidth. It's designed to do very specific stuff. And like someone said, most processing is done on earth, probes and rovers only collect raw data and decoding of it happens on earth. They just need to control the thing, collect data and send it to Earth reliably. That's all it matters. The rest happens here.

[Zodiark1593]

9 minutes ago, Mihle said:

Curiosity had 2 MP camera, but the Perseverance actually have a 20 MP camera. Cameras are a thing they have updated.

 

I am guessing the CPU is the exact same one as in Curiosity.

I can definitely see cameras getting updated. While an error in a landing calculation would be very bad, an error in a pixel or three is negligible. Though there’s almost certainly some DSPs or image processors to handle the data, as I’m fairly certain a 200 MHz PowerPC 750 would have it’s hands full. 

[JovanD]

You can do a lot of useful  stuff when you don't have Windows running

 

[igormp]

15 hours ago, Fabioo said:

Not to mention the storage, I'm no space explorer and have no idea how it works, but since just the landing phase was expected to register more than 28,000 images (source: https://www.researchgate.net/publication/346537343_The_Mars_2020_Engineering_Cameras_and_Microphone_on_the_Perseverance_Rover_A_Next-Generation_Imaging_System_for_Mars_Exploration page 31) doesn't that storage seems a little bit odd?

Many people explained about the raw CPU power and its coprocessors, but another fact is that more RAM, ROM or flash means more bits than can go bad in some way, and you definitely don't want that. 8gb of storage is already A LOT, and transmitting stuff from earth to mars and back isn't that fast anyway.

[nbrowser]

18 hours ago, Eigenvektor said:

The rover mostly just has to collect data. All of the heavy processing will happen on Earth.

 

And yeah, it does have different requirements:

• Be as reliable as possible
• Use as little power as possible
• Be resistant to space, radiation etc.

Using bleeding edge technology is no good if you need something that is ultra reliable, because you haven't had decades to fix all the kinks. And smaller process nodes are much more susceptible to stuff like radiation and high energy particles. You don't have a magnetic shield on Mars like you do on Earth.

Yep, they used proven parts that are known reliable in the long run...once they sent that thing into space, there isn't a "Mars Rover Spaceside" service provider to fix it. They gotta build with what you said as the utmost importance, reliability.

[Morrie Sells Wigs]

5 hours ago, Sakuriru said:

I actually wrote a paper on the subject of software engineering design choices of interstellar probes, which is a similar problem.

 

Each time one of these devices is constructed, they face a unique combined problem that they're not serviceable, have long development pipelines, and have requirements that can change frequently. This makes it particularly challenging from the full software packaging perspective. Normally with these kinds of designs a more waterfall or v-shaped SDLC is followed to ensure maximum reliability. Hardware is no exception and often the entire thing is written from the ground up. In this case Linux was chosen as a base, but really you want to narrow your scope to a specific spec and build to that spec. When millions of lines of code have been written or altered the last thing you want to do is change the amount of RAM in the system and hope it hasn't broken anything down the line.

 

Really, it's just the kind of turnaround that you would expect from an extreme reliability requirement.

  

You didn't give an opinion, you made an incorrect statement.

Incorrect, according to you.

[LAwLz]

The best thing about this thread is that OP thinks NASA would need help from LinusTechTips to build a more powerful computer. As if

1) NASA doesn't have the funding and expertise to do it better themselves.

2) Linus would know how to build a computer that is going to Mars.

 

OP, you are either dramatically overestimating Linus' knowledge or dramatically underestimating NASA's resources.

[Morrie Sells Wigs]

18 minutes ago, Sakuriru said:

According to NASA, actually, I'm not making up information. Your position doesn't make a ton of sense; it's not like what they did isn't conceivably possible.

According to N.A.S.A...lol.

 

Yeah, because they never lie.

 

Seriously, some people are so blinkered and they need to stop believing everything the nice lady on the mainstream news tells them and they need to start thinking for themselves. 

I'm not actually saying it is fake - I'm saying that we can't possibly know if it is fake or not as we have no way to check for ourselves.

 

For example, I can verify that my tv exists because I'm sitting in front of it watching it.

 

I can verify that my dog exists because it is lying next to me.

 

I cannot verify that the perseverance rover exists.

 

We can trust N.A.S.A, and assume they are being 100% honest, or we can be distrustful...and we can trust the mainstream media or we can be distrustful. 

 

It's about how you see life.

 

Whatever else happens, however, we should, at the very least, be able to trust our own common sense and stuff we have researched ourselves. 

 

Anyway, like I said, I was half joking.

[LAwLz]

1 hour ago, Maury Sells Wigs said:

According to N.A.S.A...lol.

 

Yeah, because they never lie.

 

Seriously, some people are so blinkered and they need to stop believing everything the nice lady on the mainstream news tells them and they need to start thinking for themselves. 

I'm not actually saying it is fake - I'm saying that we can't possibly know if it is fake or not as we have no way to check for ourselves.

-snip-

The whole "we can't see it so we can't know if it's real or not" is kind of a bad argument. I can't see your brain but I am fairly certain it exists in some capacity. I haven't seen the mars rover but we do know that NASA has sent out things into space before, otherwise we wouldn't be able to get satellite images of Earth and things like GPS working.

 

At a certain point it makes more sense to just believe in something even if you haven't seen it with your own eyes.

[tikker]

39 minutes ago, Maury Sells Wigs said:

According to N.A.S.A...lol.

 

Yeah, because they never lie.

 

Seriously, some people are so blinkered and they need to stop believing everything the nice lady on the mainstream news tells them and they need to start thinking for themselves.

It's always good to be sceptical and to form your own opinions about things. Science communication is a difficult field in addition to that, especially when it comes to higher level physics or extraterrestrial travel, that aren't among the average Joe's day to day activities.

39 minutes ago, Maury Sells Wigs said:

I'm not actually saying it is fake - I'm saying that we can't possibly know if it is fake or not as we have no way to check for ourselves.

I can never really grasp this type of argument. Mostly because it's often an argument of incredulity and unwillingness. The entire point of scienctific literature, is to show your result, argue why you think it is correct and write down your methods so that anyone who has your data and follows your method should obtain the same result. To tie this to the real world, if I give you a manual to assemble a bicycle and you follow that manual, you'll end up with said bicycle. If you end up with a different or non-functional bicycle, you inform the community you found something different.

 

This happens all the time in science from "debunking the CMB" a few years ago, which turned out to be dust in the Milky Way (IIRC), to the recent discovery of phosphate on Venus which lead to two opposing camps. One believing the detection and others argueing it wasn't detected at all. This could be done because they weren't simple claims, they were well documented and argumented publications.

 

My point being here: the beauty of science is that it's out there for anyone to repeat and confirm for themselves. Many have done so and that is why they have become established as current truths. We believe them, because no better explanation has been offered so far.

 

Our current scientific method is built around doubt and being wrong. Science doesn't (and can't) prove truths, it merely accepts as true that which has not been proven false. To ascertain the truth of something, you set the circumstances under which your proposition should hold and you start trying to disprove it. If you can't, you must accept it as true until you either can proof it false, or properly argue a different solution to be more correct. You could in principle see our scientific method as a fallacy of its own: the appeal to authority argument that a scientist is a reliable source of information.

39 minutes ago, Maury Sells Wigs said:

For example, I can verify that my tv exists because I'm sitting in front of it watching it.

 

I can verify that my dog exists because it is lying next to me.

 

I cannot verify that the perseverance rover exists.

I cannot verify that your TV or dog exist. I don't even know who you are. Does that mean they don't exist? If you give me your address and a way to get there however, I can. It's the same for the rover. If you take a rocket to Mars, you could verify it. Slightly simpler for current times, you can use the Deep Space Network to listen for signal from Mars. If Perseverance is actually there, you'll receive the signal it is sending back to Earth. It's perfectly verifiable that the rover is there.

50 minutes ago, Maury Sells Wigs said:

We can trust N.A.S.A, and assume they are being 100% honest, or we can be distrustful...and we can trust the mainstream media or we can be distrustful. 

 

It's about how you see life.

Instead of being distrustful, we should live with a healthy sceptiscism. If you told me the atlantic ocean is blue I might be sceptical, because all my local bodies of water are orange, and I would try to find some (scientific) literature on it. A quick Google Earth would verify they are indeed blue, and I would find out that it is blue because water absorbs most wavelengths aside from blue. This would lead to the question why my local "water" is orange etc.

1 hour ago, Maury Sells Wigs said:

Whatever else happens, however, we should, at the very least, be able to trust our own common sense and stuff we have researched ourselves.

Common sense is a tricky one when it comes to feats like this. In some way landing on the Moon or other planets defies our current common sense, because it's not a common thing to do. Sadly "researched ourselves" has become tricky with paywalls for scientific journals (plenty available as preprint though) and more importantly clickbait articles that can misunderstand the actual meaning. Something that happens often is "we might have found this odd thing" getting turned into "Scientists discover X, disproving physics as we now it!".

 

NASA, ESA, JAXA or whatever space agency know what they are doing. They have many talented engineers and scientists working for them and they have stuck around, because they have gained trust. That trust they have built over time with their results. We are told about successes, but also about incidents, including fatalities. If they continously published garbage, built rockets that exploded all the time they would gradually lose that trust.

[PlayerLorenzo]

The computer inside is not designed to play games or browse the web. It is designed to withstand radiation, provide instructions to navigate the Rover, and be reliable and fail-safe. And 256 MB of RAM is good if the only thing you're doing is navigating Martian terrain . Oh, and don't expect the thing running a complex OS such as Windows or even Linux, it probably runs a very basic proprietary real time OS that is designed to be reliable and fail safe, along with being responsive even by the microsecond.