Slashdot Mirror
Mars Rovers Update
BoldAC writes "CNN is reporting that engineers will upload a software hack to decrease the recent power drain plaguing the rover Opportunity. The hack works by reducing the power supply to a poorly functioning switch." p3tersen writes "Opportunity has photographed a blue martian sunset (it's blue because of the optical scattering properties of dust in the martian atmosphere). In other news, the rovers are beginning to experience power supply problems due to the accumulation of dust on their solar panels."
NASA should have installed wiper blades on the solar panels.
"The hack works by reducing the power supply to a poorly functioning switch."
You know, they could just tell the rover to use its ACPI functionality and go into standby and spin down its hard disks....
print 'Hello world!';
http://compbrain.net
Is there some reason why these rovers do not have a fan to blow away the accumutating dust on the solar panels?
The weight penalty should be offset by being able to work longer.
Or is the dust sticky? Maybe something akin to a wiper?
- - - - - - - - - - -
I am a programmer. I am paid to produce syntax not grammar. Deal with it.
The problem was probably caused by some incompatibility in the Java system
Repeat after me.
There is NO Java on the rovers. Java is used on the ground to process the results.
Idiot. Enough has been posted on this site about where Java is being used.
The dust problem is just indicitave of how difficult it is to plan such a complex mission like going to Mars. Until we do this a few more times and figure out a few more things, a manned mission will have to wait.
Java is only used on the ground. They'd be stupid to use it in the flight software.
That isn't a slight against Java. It's just that they need real-time software, which can't be had with an interpreted language (even if it is only interpreted from bytecode).
How fast do you think a 20mhz processor could run a Java app?
A nice photo from the Cassini mission.
The blurb fails to mention that seasonal changes on mars are resulting in less sunlight per sol. That is one of the main power issues.
Dan East
Better known as 318230.
...to have the dust wiped off the rover's solar panels, it has many disadvantages. It has been discussed widely here, even by rocket scientists. In short, it is extremely difficult to come up with a "cost-effective" (from several viewpoints) mechanism. An interesting fact is that the cost of operating the entire mission is around US$ 3m a day, and that must also be considered when determining how long these wonderful bots rover through Mars unveiling its mysteries.
The revolution will not be televised.
1. It's going to weight more.
2. It's another potential failure.
3. IF it fails, it can cause other things to fail (say, for example, a switch sticks ON and it drains the battery)
Not installing a wiper or other device to clear the solar panel wasn't an oversight. They made a (probably) well-informed decision not to install such a device. I think the progress so far is remarkable and should be commended. Hopefully they've learned a lot and can make improvements for the next mission.
Hindsight is always 20/20...
The big problem is that the dust is so fine, it'd be very difficult to wipe it off with anything akin to a windshield wiper. You might remove the dust, but the grit would scratch the glass, eventually causing enough opacity that the panels would eventually be rendered useless.
One thought I had was to gradually apply a charge to the solar panels and then suddenly apply an opposite charge, causing the dust to be repelled from the surface, to be carried away by the Martian winds.
I've no idea if it would actually work or not, but it seemed an elegant solution that didn't require any moving parts.
Guaranteed! This comment 100% Anthrax free!
It just plain makes sense, when you think about it.
Visit CryptoGnome in his home.
Some important (I presume) Sun person made the Java on Mars claim at a recent Sun Tech Day. She was quickly corrected by another (more important) Sun guy but it's easy to see how people are getting confused if some Sun sales people are spreading this rubbish.
Why did NASA stray from 'nuclear' batteries, like they've used with the Pioneer, Galileo, Voyager and Cassini missions? Those could power a rover for years.
And what's stopping them from making a way to keep the panels clear? This is what contributed to the end of the Pathfinder mission...What is it about solutions to this problem that make them so difficult to implement?
Wipers add an extra mechanical system to worry about, but what about static fields? Maybe there could be a way to attract the dust to a specific area while keep the the panels clear?
Use a Vibrator! No Really!
I've been reading all the other posts, Every idea from peelable plastic sheets to fans..
Just attatch a vibrating motor to the underside of the solar panels. When it's time for them to get clean, just raise them to a 90 degree angle, turn on the "orgasmotron vibrating motors" and shake the dust off?
Maybe I watch too much pr0n, but I'm sure that would work for the heavier dust. Especially since there was an earlier comment on how the engineers purposefully drive these things over rocks to shake off the larger dust particles.
One more thing, movable solar panels can track the sun, and give better light collecting efficiency than ones that just sit there stationary.
--toq
How does testing the software determine the heater is going to malfunction?
Given CNN's lack of technical depth, for all we know it could be a command to the rover to tell it not to turn on the heater anymore. Either way, they designed the rovers so they can fix them while they're in space, which is pretty good.
Given the limited amount of storage on the rover, it's a higher priority to make sure the upgrade process works, and that it is possible to fix stuff with software, than to make the rover fix stuff automagically.
Need a Catering Connection
I'm sure a bunch of half-baked ideas from random yahoos would get them much further than anything their silly engineers could come up with.
--
the strongest word is still the word "free"
Once the robots find water, they could spray it on the solar panels to clean the dust off...
A thread attached to a corner of every plastic sheet running diagonally to the opposite corner and an electric motor that activates when the sheet is dirty.
Stack 10 sheets of this and voila, lifetime of the rover multiplied x10.
Conceptually, this is a great idea, except for one problem:
Every layer of $whatever you put on the panels attenuates some of the light striking the panels. The sunlight is also that much dimmer there (at the very least by the inverse square law of distance from the sun, if not also because of atmospheric conditions), so every single watt-hour those things can capture is critical.
Of course, to compensate for the thin film layers, they could have made the solar panels bigger - but that adds launch weight... not to mention the bigger solar panels would make the whole thing more top heavy and likely to tip over due to wind or ground obstructions, meaning you'd want to add size and wheelbase to this thing, meaning you'd need more solar panels... Do we see a vicious circle yet? [grin]
Fire and Meat. Yummy.
More useful, and what appears to have been implimented, would be a system where they can easily upload fixes/patches when un-expected errors appear.
I suggest a solar panel Zamboni! If for no other reason that Zambonis are cool and a space Zamboni would be a magnitude cooler. Imagine the great PR if NASA could send back shots of there Zamboni working on another planet.
Papa Legba come and open the gate
I have to say though, despite being an extremely short video clip, it is one of the most awe inspiring things I've seen in a while. Think about it. We just viewed a sunset ON ANOTHER PLANET. I can just imagine an art gallery featuring nothing but pictures of sunsets on other planets. As much as I love our planet Earth, I hope the day comes when I'll be able to stand on Mars and watch this for myself. The beauty of the universe is infinite, but every now and then a little piece of that beauty finds its way back to Earth, and we experience this beauty, and smile a little, not quite realizing the magnitude of what has just occured.
Buy Steampunk Clothing Online!
Why did NASA stray from 'nuclear' batteries
Because the feel-good environmentalists have heart attacks when you mention the word nuclear. The risk is very low and the potential benefits to science are very high. The amount of nuclear material required is not large.
There are some problems with output from nuclear batteries, and I don't claim to be a rocket scientist - but surely it could be made to work.
I'll be laughing my ass off when oil triples in price and people are rioting - we'll see how long the nuclear boogeyman scares people then.
..don't panic
I think the main issue with Java is not the processor, but several other qualities that, while highly desirable for a general purpose language, are unacceptable for real-time systems (garbage collection is one of them).
:)), and according to my Concurrent and Realtime Programming in Java lecturer (Andy Wellings, who sits in the Technical Interpretation Committee for this JSR), NASA is quite interested in this as an option for its future missions to Mars.
This is why there's been work for a real-time version of Java for a number of years now (see here). Incidentally this is what kicked of the JCP (it's JSR-001
The revolution will not be televised.
this is slightly OT, but "real-time" in engineering circles does not mean "really fast." it means that there is a guaranteed response from the system within an specified actual time frame.
i.e. I need a real-time OS & software stack if my rocket control algorithm needs the data from, say, a serial port altimiter within the next 20 milliseconds or else. if you cant get the data within the specified timeframe then the results are useless. the system will not accept requests that it cannot "guarantee" to fulfil from a system resource standpoint. (you have to watch your multitasking, swapping and other kernel-level tasks to achieve this)
so you could have a 20 mhz "real-time" system, as long as it's response was guaranteed by the OS within parameters for what you are doing (and you would program with those guaranteed response times in mind.) Conversely, a 20 Ghz system may not qualify for real time, if the OS pre-empts your rocket control task and decides to swap for a few milliseconds too long, or context switches to another thread just when you needed to adjust a control surface...
when you hear about people hacking linux for real-time work, they are not making it go faster (though that's always nice), they're making it work predictable.
This makes things sound worse than they actually are. They are not beginning to experience power supply problems -- they are simply getting less power than they were when it first landed, and they are taking some steps to operate more efficiently.
From SpaceFlightNow, in the report for THURSDAY, FEBRUARY 26, 2004 2215 GMT (5:15 p.m. EST):
From the Reuters report:
What the NASA official (Richard Cook) actually said was: "The vehicle is continuing to perform fine but we are starting to modify our daily process to respond to the decreasing power."
Both the dust accumulation and the decrease of sunlight were anticipated. The lifetime (designed to be 90 days) of each Rover is determined when the batteries can no longer be charged enough to survive the cold nights. Spirit is already 54 days into its 90-day "death sentence".
HCG 50a = 2MASX J11170638+5455016
11h17m06.4s +54d55m02s
Maybe it would be better if we tested the software more on earth and had all the bugs worked out before we sent it up.
They had a small time-window to get a lot of stuff ready. Because of the politics of funding, often they only have about a 4-year lead to put it all together. Plus, Mars is only in the right position about once every 2 years.
Table-ized A.I.
You can't blow them off with opposite charge easily. These particles stick to the (non-conducting) surface probably because of Influence. If the surface would be conducting they wouldn't stick because they would get discharged. This is similar to the dust particles that adhere to CRT screens, just here its the screen thats charged (by constant electron bombardement) and the dust particles get influenced. The fact that the surface is non-conducting will also hinder you bringing any kind of charge to it. Another problem is that the charge might be randomly generated and so half positive and half negative, so you could only blow off half of the dust. There is a possibility of making transparent surfaces conducting (coating them with metal films), but this reduces transparency quite a bit, and I suppose the mechanical properties of a metal film are far inferior to whatever they used (you want this surface to be hard).
The updates include improved driving software, something that only evolves after some hands-on time on Mars itself. The patch to strangle the errant heater is pretty drastic - it shuts down some primary circuits including the internal clock, which has the -side- effect of cutting off power to the heater overnight. The rover will wake up the next day only when it starts to get some sunlight on its' solar cells - not by an alarm clock as it does now. The heater will still run once the rover wakes up the next day.
======= ~\_/~\_O Burmese
It's a neat little $10 million 50 kilo unit. The best part is that a software upgrade improved the stability 10x. Hopefully there'll be some pictures soon.
One line blog. I hear that they're called Twitters now.
I'm sure some guy with a squeegee will jump on it in no time.
The APOD site had this picture of a "named" rock a couple of days ago.
Use your head, can't you, use your head,
You're on earth, there's no cure for that - S. Beckett
Arthur C. Clarke wrote a story reminiscent of this, set on the moon. Two explorers in a dusty area, and some of the dust sticks to their faceplates. Wiping the dust off builds electrostatic charge, attracting more dust, so they're soon blind.
***SPOILER ALERT*** (and suggestion)
Since rubbing causes static electicity, they rub their faceplates together. One charges one polarity, the other the opposite. So one faceplate comes out even dustier, the other clean. The explorer with the clean faceplate can lead the other back to the vehicle.
Actually I always though static electricity came from rubbing dissimilar materials, so I wouldn't expect rubbing two identical-material faceplates to do squat. But there may be a lesson here. If the primary problem is really electrostatic, might there be some sort of electrostatic solution? (on future rovers) The most extreme would be an ion-wind generator with the 'benign' (dustwise) polarity attached to the panel. Another might be a charged wiper blade. I'm sure there could be other simpler electrostatic-based solutions.
The living have better things to do than to continue hating the dead.
I think its' safe to say that the approach to programming that the rover programmers use has little in common with that of some large software companies. "Getting it out the door" (ie, launch it within the possible window) involves a firmer deadline than any MS programmer -ever- faced. They actually uploaded a new version of the driving software while the rovers were enroute to Mars.
======= ~\_/~\_O Burmese
RAT is short for "Rock Abrasion Tool", meant to remove the upper layer of a very hard rock.
In the same way you'd not be keen to use a RAT to brush your teeth, you probably would not wish to use the RAT to clean a transparent surface of a solar panel. In fact I think you may have just given some poor engineer at NASA a heart attack just by suggesting the RAT come near the solar panels!
"There is more worth loving than we have strength to love." - Brian Jay Stanley
Ok, hear me out.
Cost aside, there most likely is a way to greatly extend the life of a rover.
From the noises NASA has been making, there will be a series of unmanned missions to Mars before an attempt to send humans will be made (I don't necessarily agree with the premise of sending humans to Mars).
NASA has said the limiting factor is power, because of the dust accumulation on the solar panels.
Let's say that the cost of implementing a way to eliminate the dust doubled the cost of the mission (probably would be less). If dust were eliminated, then the rovers could operate until the batteries could not longer hold a sufficient charge to do science.
What I'm getting at here is politics. A solution could be engineered to greatly extend the life of the rovers. But that would result in a reduced number of missions and *less funding*.
Ok, I've removed my tinfoil hat. Comments?
it's blue because of the optical scattering properties of dust in
;)
the martian atmosphere
That's Scullys Xplanation.
Mulder says different.
Or it's because NASA's mission faking division forgot to photoshop the
images before releasing em.
siggy played guitar
It's called fines not dust, you insensitive clod!
That's like calling dust gravel, jeeze.
(with apologies to KSR's Red Mars)
Working toward a usable PDA environment in the spirit of Newton OS: Dynapad
On Soviet Mars, the Pink sunsets are BLUE!
Why slashdot? Why not?
My understanding is that a variety of dust removal techniques were considered. The final decision was that an effective dust removal aparatus could be added, at the cost of any one of the instruments. The scientists decided limited life with more instruments was more useful, and so the dust removal system was left off.
So why nukes for Viking, and none for MER A & B?
1) Viking had money. Sure, NASA was getting into a budget hurt locker by the time the missions made it to Mars in '76, but the money was there when it was needed during the planning and construction. The landers got the kitchen sink, and the biggest Titan II launchers then avaiable to get 'em going. By contrast, the MER team had to make sure their package was not much heavier and absolutely no bigger than Pathfinder. The planetary missions are bastard stepchildren to a NASA which is mandated to keep the Space Shuttle and ISS going on an inadequate budget, even if it all went to the manned space program.
2) Three Mile Island, Chernobyl. Hadn't happened yet, so the no nukes crowd was still the wacko fringe during Viking. Compare to the fuss made over Cassini before launch and while making a gravity-assist Earth flyby. "200,000 deaths!" "Dump it in the Sun!" In general, people have mellowed out a bit, but the PR angle makes a good excuse when one doesn't have the money to gold-plate a mission, anyway.
Luke, help me take this mask off
Who okay's these stupid ideas at NASA? What dumbass is running the show there now, and how many astronauts did he have to barbecue to float to the top of that bowl?
It's called nuclear power folks, and you can use it to create extremely reliable electric generators with a shelf life on the order of decades, and extend the life of sattelites and roving projectiles for years, not days. Day or night 'til the wheels fall off.
In fact, if you use said nuclear/electric power generators to power ion drives, you can use less than half the propellant to get to [insert planet here].
I don't get it. When did NASA become a jobs program for morons? I mean, morons have to make a living too, but shit, can't we let them execute water fountain projects instead?
They're so incredibly smart, in fact, that they don't even need to convert metric measurements to the archaic system they insist on using.
Don't knock US/SAE measurements, there's a good reason they've stayed around.
For scientific analysis, without question, Metric rules.
But when you're actually building and working on things, most of the time a 10% tolerance is good enough. As a result, usually you can stick your thumb across something and say, "Yup, that's an inch - close enough". The base units are more intuitive, although admittedly the interconversion between units is a bitch - but conversions are more common in analysis than construction/maintenance.
My perspective here? Canada went Metric in 1976. I grew up in Metric. I went to school in Metric, fuelled up my cars in Metric, got a set of Metric wrenches when I was a kid, etc. Heck, you wanna know Metric inside and out? Try taking an engineering degree in Canada!
And yet, I know I'm 6'4" tall, 185lbs. I don't know in Metric.
Every time I work on a car, I want to know first, Metric or SAE? (And I don't mean the speedometer, they've all been Metric in Canada since 1976.) Not because I care which wrenches, sockets and feeler gauges I bring, but because I like working on SAE much more.
Why?
I've had more cars with Metric fasteners and specs than I have SAE, and yet, somehow, I can still just put my thumb across a bolt and know, "Hey, that's not 1/2", that's 7/16"!" Why can't I do that with Metric? I sure which I could, especially since I've got more experience with Metric.
Fire and Meat. Yummy.
A lot of people are discussing ideas to remove dust from solar panels. Something people are forgetting is that there's a downside to having the mission last longer than originally expected: it costs more. JPL designed the rovers to last about 90 days, and NASA gave them enough money to pay the hundreds of engineers and scientists it takes to operate the rovers for 90 days each. They of course have the option of extending the mission for longer than 90 days, but the money to pay for that extension will come right out of NASA's Mars program, which means less money for future Mars missions (including Mars Reconaissance Orbiter 2005, which is already well under way and needs every penny it can get).
I attended a seminar last week with David Des Marais of NASA Ames reserch center. He noted that, while the panels were covered with dust, the rovers were still maintaining greater than peak power. He specifically stated that the rovers were shunting power because they couldn't store anymore. They suspect that the rovers will far outlive their contracted period.
There are many aspects of Java that mean you can't guarantee a response within a certain timeframe. Garbage collection, for example ... there isn't necessarily a 1:1 instruction translation of the JVM's bytecode, and therefor you can't know beyond a shadow of a doubt that a certain operation will get done in a certain amount of time.
you can get real-time behavior out of an interpreted system, if both the interpreter and underlying operating system are designed for it. For Java, GC and instruction translation just have to have that as a design goal. As a matter of fact: here's some information on real time Java:
"With the recently released Real-Time Specification for Java (RTSJ), developed through the Java Community Process by the Real-Time Expert Group, the real-time embedded software developer will be able to use the Java programming language in applications where predictable/hard real-time behavior is a must."
that being said, java probably would run too slowly for most applications on a 20 mhz cpu. But being interpreted or having GC are orthogonal issues to real-timeness.
Does anybody have a link to Mars Rover science-in-progress? We are getting all sorts of operations/engineering reports and neat photos, but almost no science reports. What have the spectrometers found? Does the rock chemistry correspond to known minerals on earth or is it new? What ever happened with the briny mud speculations a few days back? I suppose there is a methodical plan to analyze and release papers, but it sure would be fun to at least know the basic composition of those sphericles for coffee break discussion.
"...while history is usually explicable it is often irrational" --Roger Spiller
So I just happened to be lucky enough to get front row seats (I work as a sysadmin in the physics department here) to a talk by one of the people on the JPL team that works on the lander, and he mentioned this earlier. It's a bit more than a little hack to the software because it involves changing out the operating system and turning the rover completely off during the night to avoid power drain. What the fellow talking about it mentioned was that there is the possibility that the rover wont actually turn back on after the update, leaving a $400 million piece of junk on the surface of mars.
The reason for the update is needed because there is a heater on the rover that defrosts the probe that allows them to take samples from the rocks and such--which wont turn off anymore. This might not be a problem except that it puts an excess power strain on the rover, meaning that its useful life is greatly diminished. So essentially this hack means turning everything off at night because they can't switch off just the heater.
My goodness, not even Mars is beyond the reach of Disney!!
IIRC the 3M company makes some polonium-powered dust-neutralizing brushes-- something about the beta rays charging the dust particles so they repel each other. Any idea why this couldnt be done to the solar panels?