Slashdot Mirror
New Software Stops Mars Rover Confusion
MattSparkes writes "The Mars rover Spirit used to get quite confused when it came upon a rock. Because it could only plan routes of a metre or two it couldn't understand how to navigate around large objects, and frequently used to rock back and forth for hours trying to figure it out. NASA have written new software called D* for the rover Opportunity, which should allow it to autonomously plan routes up to 50 metres long. The new software still won't be able to avoid sand-traps, though."
actually controls the rail gun which will be used to 'eliminate' obstacles
Do not try to read the dupe, thats impossible. Instead, only try to realize the truth
What truth?
There is no dupe
..and I can't avoid sand-traps either!
I'm guessing the name is meant to suggest similarities to A*, but you think they'd mention that in the article.
Ben Hocking
Need a professional organizer?
They fixed the navigation issue, but I bet the rover still hates Mars.
You know, NASA sometimes has a hard time figuring out the difference between Meters and Miles. Now, you are asking them to figure out the difference between Metres and Meters.
but I have to wonder why they didn't send it up there with the ability to plane long and short treks?
The more I learn about (semi)autonomous robotics, the more impressed I am with what the Mars rovers have accomplished already and the people who built/maintain/program them.
My forays into hobby robotics tells me that this isn't as easy as it first appears it might be.
Support NYCountryLawyer RIAA vs People
Every time it would figure out a way around the rock, Marvin would move the rock.
Ha!
This story is a perfect illustration how stupid the current generation of state-of-the-art AI is (or lack of).
Anyone who ever said `there aren't any opportunities in IT', try to solve this problem!
"If anything can go wrong, it will." - Murphy
...way back in the original Warcraft. Looks like JPL needs more game programmers to help with their "pathfinding" algorithms.
Well if its taken this long for NASA to figure out a good routine for pathing then I suppose I shouldn't be so down on Obsidian for pathing being so terrible at it with your henchmen in NWN 2.
Is anybody else amazed at the apparent simplicity of a lot of the problems facing the rover? I don't say this to criticize the NASA guys and gals, I'm sure if I had designed a rover would never have done as well up on Mars as theirs, but it always surprises me how simple a lot of the problems they face seemingly are. Of course, a simple problem here is a big problem when you're trying to do it at 35-250 million miles away.
Gentlemen! You can't fight in here, this is the war room!
as the Rover left its pitching wedge back at Canaveral.
Oh, yeah, it's not easy to pad these out to 120 characters.
Come on slashdot... railguns... eliminate... its a joke. :-)
Pete/Petri "damn, my chainsaw is clogged with 1's and 0's again." --clyde
I believe that in many English-speaking countries, collectives (rock bands, corporations, etc) are treated as if they're plural. Maybe government agencies follow the same rule.
The fact that these rovers are still working far beyond the length of time they were intended to last is remarkable in itself. That they can still be programmed to change their routine is something I find remarkable. Yes, they are simple tasks, but it is still quite an achievement.
Dude, welcome to the English-speaking world.
/ grammar/learnit/learnitv163.shtml
http://www.bbc.co.uk/worldservice/learningenglish
Cheers.
--I'm so big, my sig has its own sig.
-- See?
In British English this would actually be correct. In American English, not so much...
Protect your browser with the Force Safe Search add-on
Game maker Ubisoft has filed a lawsuit against NASA, claiming it stole code originally used to control GRAW AI teammates for use in the Mars Spirit Rover.
I believe that in many English-speaking countries, English is utterly misunderstood. England is one of 'em :) I mean hell, they're the guys who changed from "Recognize" to "Recognise" simply because we Americans favored the -ize form. They also can't spell Aluminum (as per the wishes of the person who got to name it) and boil everything into submission. But on the plus side, they will be happy to tell you that they hate the French more than we do here in the US, so at least we have some common ground.
Nonetheless, pluralizing the name of a composite entity like NASA may seem to make sense at first, because let's face it, talking about NASA as a single entity misses the point that it is made up of people; but on the other hand, and in fact this is the hand I'd like to slap people with when they pluralize the names of these composite entities, is that it is in fact often meaningful to talk about the organization without talking about individuals. In those cases it makes no sense whatsoever to pluralize.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
Slashdot Burying Stories About Slashdot Media Owned
Tiger recommeded a 7 iron; get him a copy of the Rover SDK...
It must have been something you assimilated. . . .
Duh, it converts the rover into a router so mars doesn't blow through it's IP space. See the earlier story on the intraplanetnet...
and how much money and time did you donate to worthy causes last year.
Undetectable Steganography? Yep, there's an app fo
It seems since the first year, the rovers have been more useful as a means of testing robotics software than as a means of studying geology.
When the problems on this planet make it uninhabitable, it'll be nice to know what Mars is like before we try to move there.
And in comparison to the federal budget for social security and the military (combined something like 85% of the total budget), a few million for a Mars rover is a drop of the bucket.
It's really a question of values, and what you consider to be "problems" that are more important. That's a subjective judgment, and you can't really say it's "wrong" to spend money exploring Mars.
Some NASA software engineer reading slashdot is probably shitting his pants right now because he wrote the config line:
MAXSEARCHLENGTH: 50 yards
"Can I come home now? When I agreed to go to Mars I was told it was a tropical paradise and I would only have to stay for three months. It's cold and dry and I haven't seen a single palm tree, not one! The software update was a big help but I'd rather have a ticket home." Signed Spirit Rover. "I miss you guys. By the way who won the last three American Idols? Reception sucks up here."
Pioneer 10 & 11 came out of Nasa Ames.
Yes, D* has been used "live" for the first time.
However, both rovers received a fresh load of mission s/w a couple of months back which enables a variety of fabulous new functions, including "go and touch" (as opposed to the original "touch and go") - go and touch enables the drive planners to instruct the rovers "move 12.4 metres forwards, turn 30* left, forward 70cm, approach the rock in front of you, deploy the IDD (robot arm holding a variety of instruments, spectrometer, close up camera, the RAT (grinder) and brush, etc; deploy the Mossbauer spectrometer, take reading in situ for 18 hours".
It also enables them to build their own route maps. One problem is that on featureless plains, it needs landmarks to assess how far it's travelled -- thus the newly developed "drunken sailor" manouever, designed to make clearly visible tracks that can be used to triangulate the on-board navcom. thing.
"None are more hopelessly enslaved than those who falsely believe they are free." -- Goethe
The oscillation problem looks quite trivial to solve in theory. If they can set a target to the rover, that means
they can use an informed search strategy using heuristics like offline A* http://en.wikipedia.org/wiki/A*
or online LRTA* (Learning Real Time A*), which will solve any oscillation problems. If they have high def images
it would make it even easier. I assume they use informed search strategies, since uninformed search strategies
could lead to exponential search trees.
Even with uninformed search strategies though, you can still avoid oscillations (repeated states) by
using open lists and closed lists. That is, you put a visited state in your close list so you dont visit
that state again. You can easily reduce the number of states by setting the world into a grid map (2 dimensional array).
This happens all over the place in games. It's basic staff in AI, but I'm not sure how it works in real life.
for (appropriately-named) OnStar, this is it! "Hey, there's a huge rock in front of me, how do I get out?"
This 'information' was 'officially' released November, 2006. All technical and stuff, bristling with pride about Spirit's 'new' abilities thanks to software 'D'. I'm left wondering how they got onto Mars with such rudimentary navigation in the first place, but NASA explained that due to the 20 minute lag, both rovers weren't allowed to map out more than a meter or two, and this somehow led to a 1.5 hour stutter-shuffle or something that meant they were simply not equipped to handle large objects. Was there a part of the design spec. that said there was not to be a requirement for either of the rovers to need the ability to navigate around 'large objects'...? If so, who the hell came up with that joke...?
Ok, so I figured I'd do a bit more digging to see what, if anything, I could learn about rover navigation pre-2006.
As found on the NASA website, today:
February 9, 2004
NASA's Spirit rover has begun making some of its own driving decisions while its twin, Opportunity, is presenting scientists with decisions to make about studying small spheres embedded in bedrock, like berries in a muffin.
Both rovers are on the move. Late Sunday, Spirit drove about 6.4 meters (21 feet), passing right over the rock called "Adirondack," where it had finished examining the rock's interior revealed by successfully grinding away the surface. The drive tested the rover's autonomous navigation ability for the first time on Mars.
"We've entered a new phase of the mission," said Dr. Mark Maimone, rover mobility software engineer at NASA's Jet Propulsion Laboratory, Pasadena, Calif. When the rover is navigating itself, it gets a command telling it where to end up, and it evaluates the terrain with stereo imaging to choose the best way to get there. It must avoid any obstacles it identifies. This capability is expected to enable longer daily drives than depending on step-by-step navigation commands from Earth. Tonight, Spirit will be commanded to drive farther on a northeastward course toward a crater nicknamed "Bonneville."
"The drive tested the rover's autonomous navigation ability for the first time on Mars."
What? Back in 2004, NASA clearly stated that 'stereo imaging' was the technology driving 'the rover's autonomous navigation ability'. That was Spirit they talked about having stereo imaging then, and now today, they say Spirit lacks anything allowing it to go ahead on its own....
So what is up with this lame story back in November where they say both poor little rovers can't tell a wall from a boulder from a breadbasket. Load of crap, my meter says....either way, NASA is pulling legs. Was someone full of shit in 2004, or in 2006? How about both times...then, then and now.
If they were in Tartarus Colles, they well might drive over a bridge :)
Stop it, stop it, it's fine. I will *destroy* you.
When Mars is at its furthest from us it only takes 21 minutes to receive a signal from the rover. (And, yes, I know it's blocked by the Sun then.) That's 42 minutes round-trip. I guess one explanation is that their first attempt failed and they didn't find out until 42 minutes later. (Alternatively, and more likely, their first X attempts failed and Mars was closer.)
Ben Hocking
Need a professional organizer?
Thank you for letting us know and even providing a link to the tech report.
Here's one that that really bugs me about online news articles: they rarely take advantage of the medium! Why the hell couldn't New Scientist provide a link to this paper? As the parent mentioned, the very least they could have done would be to mention that it was related to A*. Yes, I realize that the "average reader" (whatever that means) may not want to know such details, but why couldn't they include a simple "(details...)" link right next to the mention of D*? An interested reader could click on the "details" link and expanded text would create a brand-new paragraph describing more details of the algorithm and providing a link to the tech report. Click again, and the detail paragraph collapses and you go back to reading the basic-version of the story. Simple!
These publishers haven't grasped the power of the new media. For them, online articles simply mean that it's faster, easier, and cheaper to publish stories. Basically, they embrace web publishing for their own selfish reasons. With just a modicum of thought, they could make their products much more valuable to the consumer. But I suppose that would require them to give a damn about their work.
GMD
watch this
2004 press release
/. , no less - sigh).
2006 press release - Sol 1014 - (four months prior to being front-burner'd by the ever vigilant staff here @
A couple of Soviet probes also went past Mars.
The new software still won't be able to avoid sand-traps, though."
Guess they won't be playing much golf.
Is this really helping humanity??
Tell me, are YOU making a significant contribution to humanity? Please stop breathing, and save the oxygen for useful people.
Seven puppies were harmed during the making of this post.
What about corporations consisting of one person?
http://outcampaign.org/
It's true I tell you, feller at work's next door neighbour read it in the paper.
True, and then there's the one-man bands.
What if the Hokey Pokey really is what it's all about?
These publishers haven't grasped the power of the new media. For them, online articles simply mean that it's faster, easier, and cheaper to publish stories.
For them, it's about entertainment and distraction, not about providing information. Entertainment is much more profitable and less labor intensive. Personally, I wish those who submit articles would link to the very original source, instead of a fluff piece from CNN.
What?
Here's one that that really bugs me about online news articles: they rarely take advantage of the medium! Why the hell couldn't New Scientist provide a link to this paper?
Because science journalists and editors know nothing about the fields they report on, and frequently seem actively hostile to the facts in favour of some made-up, mythological story.
Some years ago a colleague at the university where I was a post-doc came into the lab and said, "Today the science section of the LA Times has five stories about stuff that I am either personally familiar with or have actually worked on. Four of them are all but unrecognizable. The fifth is full of things that are not true."
In my subsequent experience as a scientist I have found this to be pretty much par for the course for science journalism. I don't actually know if reporters are as ignorant as they appear, but by the time the edited work gets to the reading public it is almost always spun in such a way as to be misleading or simply wrong.
There seem to be some myths about science that are the bedrock of science reporting, and reporters or editors will distort or simply lie to ensure they reinforce those myths. Some of them are:
1) Science is mysterious and full of contradictions (see yesterdays "hot ice" story)
2) There is no point in anyone trying to understand science, it is beyond you (thus the lack of useful links in stories)
3) Scientists are either put on a pedestal or dragged through the mud. They are never treated as merely ordinary people doing a job or following a vocation.
4) The "story" is more important than the facts.
There are probably a lot more, but basically, science journalism will never be worth anything until it starts actually reporting on science rather than wasting time promulgating editorial myths.
Blasphemy is a human right. Blasphemophobia kills.
Well, I don't know if they would be considered journalists, and they're certainly not "popular" science journalists, but those in the Nature's weekly podcast and BBC's Naked Science podcast seem to be pretty good. Nature's is full of interviews with scientists who've recently gotten published in the journal, hosted by "Dr. Chris Smith, BSc MB BChir PhD". It's common for an interviewee to comment on how good the question they've just been asked is. I can't extract much information from them ("sodium channels?").
I wonder though of what quality magazines such as Discover and Scientific American are. Discover has covered questionable ideas, relating to the many-worlds hypothesis for example, where by the end of the article it's difficult to make anything of it ("so there may or may not be something there?"). Sigma XI's American Scientist is presumably good; at least it's very detailed and more interesting.
From the article: By angling a little bit to the right early on, the rover avoided the obstacle without needing to back up or make any sharp turns. Everyone worries that the skies will become a deathtrap when flying cars, driven by people without pilots' licenses, hit the market. But the collision-avoidance solution is simple if they're all flying autonomously. In 2007, it's trivial for inexpensive consumer devices to communicate with each other wirelessly. Similarly, flying cars need to broadcast their positions and velocities to all other aircraft within a few km radius. It then doesn't take much computing power to compute the slight course adjustments to avoid collisions, or even to avoid intersecting another aircraft's wake vortices. This will also eliminate "air lanes," and the fear of them becoming saturated with traffic. All aircraft will simply fly the shortest point-to-point great circle route, except when the computer tells it to deviate to avoid another aircraft, its wake vortices, or an ADIZ. Because three-dimensional airspace is so vast, it will be able to accomodate expoentially more traffic than the current "air lanes" concept.
That that is is that that that that is not is not.
Really? That's funny - I'm not the source of the two conflicting articles...surely you can find better ways of showing your ignorance than simply killing the messenger. In any case, you want to waste mod points, fire away.
Karma: STILL Excellent!
Why the hell couldn't New Scientist provide a link to this paper?
Why would any for-profit media publisher willingly provide a means for a viewer to leave their site?
Now, they could, as you suggested, create little "infoboxes" that expand to provide background on the topic at hand...but that requires so much more editorial effort, yet still generates no additional pageviews.
Basically, they embrace web publishing for their own selfish reasons
Oh, I see you got that. Sorry!
The US free market: two halves of a government-granted duopoly are free to set the market price.
And maybe it will inspire us to take good enough care of the planet we're on so that moving to Mars doesn't look attractive.
Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
Isn't it pretty hard to avoid the sand trap when the whole planet the rovers are stuck on is a sandtrap?
I said that same thing years ago, here on Slashdot, and the answers I got back were, "Why should they? It's not that hard to find." In that case, maybe it wasn't, but they should still do it. It's a concept I like to call "responsible journalism".
Lockheed did build the rovers. JPL doesn't build anything. Learn your facts before opening your pie hole.
I can use that for my roomba too.
1) C'mon isn't found in the dictionary -- I think you meant Come on
2) try and -- should be try to
3)
4) gah isn't in the dictionary and I'm not even sure what that is supposed to mean
5) I am absolutely sure your punctuation and your sentence structure are incorrect, but as I am not an expert on the English language, I won't even attempt to correct your grammar.
It's kind of ironic that you seem to demand the King's English on such an informal forum, when you don't use it yourself. I'll bet that you also use words like irregardless and alot.
Do you even know what it means to be pedantic?