Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
Science

DARPA Grand Challenge Updates 234

Posted by CmdrTaco from the coming-right-up dept.
GraffitiKnight writes "After only 1 team managed to successfully navigate the DARPA Qualifying course, DARPA has rewritten the rules to let almost everyone compete. Wired has the story, which also mentions rumors that the race will run to 150 miles, much less than the original plans of 210 - 300 miles." Here is some earlier Slashdot coverage of the race.
This discussion has been archived. No new comments can be posted.

DARPA Grand Challenge Updates More

DARPA Grand Challenge Updates

Comments Filter:

  • If I was that one team... (Score:3, Insightful)

    by Mz6 ( 741941 ) on Thursday March 11, 2004 @05:02PM (#8536843) Journal
    I would be pissed! To change the rules at the last minute to allow the teams that didn't do designing so well? Stupid!
    • As a member of the CMU Robotics Institute, though not Red Team, I must add that this is such a rip...well, okay, I guess we can let the others compete, but we still win, right?

    • Re:If I was that one team... (Score:5, Insightful)

      by Razor Blades are Not ( 636247 ) on Thursday March 11, 2004 @05:24PM (#8537091)
      It's only a qualifying round.

      Look at it this way - it's more fun to watch them fail in your dust behind you as you streak past them into the distance than it is to race alone.

      After all, if they couldn't qualify, but are allowed to race anyway, what chance do they have of beating the better designs ? Very little.
      • Development isn't a linear process. A team could have been 99% complete with a superior design than a team that was 100% complete with an inferior design.

        Or maybe the real competition won't test them in the same way as the obstacle course.

  • Man, I can't wait for the 8th version of this race (Score:5, Funny)

    by Anonymous Coward on Thursday March 11, 2004 @05:03PM (#8536848)
    I hear that it will be the machines hunting a prisoner or something over a 150 mile course.

  • This demonstrates.. (Score:4, Insightful)

    by GearheadX ( 414240 ) on Thursday March 11, 2004 @05:03PM (#8536850)
    ..that there's something wrong with DARPA's mentality.

    You don't accept substandard results if nobody can produce. This is something you intend to throw massive money on eventually, you'd want the would-be contractors to put up or shut up real quick.
    • Exactly. Do they think that the unmanned drones are going to end up in LESS hostile environments, whether they end up in combat or on mars? Pathetic.

      If this is an example of their evalutation procedures, and form everything I've seen it is, it's no wonder we end up with such high failure rates.

      • Basically, this isn't a real DARPA eval. It's for the cameras. First of all, the Hummer already won, go to their web site and look at who is supporting it with technology and money: Big biz military contractors. Second, DARPA does not play this way for real, this is just one big press reliese. Maybe they want to recruit new grads or something, but it sure ain't about drones.
    • almost like they are trying to make it so someone wins this year. THe money doesnt have to be awarded this year, so why lax the rules, give the teams more time and they can develop something even better.

    • Re:This demonstrates.. (Score:5, Insightful)

      by metlin ( 258108 ) on Thursday March 11, 2004 @05:11PM (#8536937) Journal
      But if DARPA pulled out because of substandard results, they run the risk of being bad-mouthed, and perhaps even losing cred.

      And the next time around they conduct something like this, not too many people would be willing to compete.

      However, this lets DARPA see more entries - agreed, some crappy ones - but a lot of good ones which are good but would have otherwise not made it. Besides, its really too early to say anything, so lets see.

    • Re:This demonstrates.. (Score:5, Insightful)

      by nacturation ( 646836 ) <nacturation&gmail,com> on Thursday March 11, 2004 @05:15PM (#8536984) Journal
      You don't accept substandard results if nobody can produce. This is something you intend to throw massive money on eventually, you'd want the would-be contractors to put up or shut up real quick.

      Then again, this is very very early in development for all teams involved. Should they cancel the event or only have one team competing? Kinda ruins the whole purpose. DARPA set an ambitious goal and, seeing that the technology wasn't quite there yet, revised the goal. Nothing wrong with that. It encourages people to participate and, by allowing more teams to actually get involved with the competition, mistakes will be made, they will learn what not to do, and the science will advance.

      Remember, they're not ordering a few billion dollars worth of equipment yet. This is mostly a proof-of-concept event to foster investment from outside parties. Start small, encourage teams to make advancements, then hold a more challenging event in a year or two. Seems like a good way to do it to me.

      • Re:This demonstrates.. (Score:2, Insightful)

        by Anonymous Coward
        Yes there is. Aren't these the guys who said we got too many entrance applications so only the big boys can play?

        To now turn round and say OK we scared of the little people so we can "open it to everyone" seems a little disigenuous at best and base cronyism at worst.

        • Re:This demonstrates.. (Score:3, Informative)

          by einTier ( 33752 )
          I thought the main reason they limited the field is because tracking 200 autonomous vehicles over a 200 mile course and ensuring they don't hurt anyone during their travels is a logistical nightmare.

          But now that it looks like the vast majority of the field won't make 200 feet, it's a whole new ballgame. When you only have to watch a football field size of land, it almost becomes pointless how many vehicles are wandering around inside it. So, why not let everyone race? It appears that most won't make it

    • Re:This demonstrates.. (Score:5, Insightful)

      by bmongar ( 230600 ) on Thursday March 11, 2004 @05:17PM (#8537011)
      I'm not so sure letting them compete is such a crime. It's not like they have agreed to a multimillion dollar contract with some company that can't produce. If the designs suck they won't win, no money out of DARPA's pocket.
      However a poorly designed bot can have a desing feature that if developed by the right people would be usefull. This lets DARPA see how some of these potential inovations will perform.
    • If you read the article you will see that one of the reasons that they changed it was people's robots couldn't navigate on the first try and needed some modifications, or something went wrong. I think that if they are really looking for a promising candidate this is a great idea. After all, they're throwing money at development -- not buying a robot.

    • Business as usual (Score:5, Insightful)

      by SuperBanana ( 662181 ) on Thursday March 11, 2004 @05:34PM (#8537211)
      You don't accept substandard results if nobody can produce. This is something you intend to throw massive money on eventually, you'd want the would-be contractors to put up or shut up real quick.

      You mean like the 'Star Wars" Missile Defense System which has failed numerous tests of increasing ease, but is being used anyway because "it beats nothing", except that "nothing" doesn't violate treaties we signed, creates a false sense of security, doesn't motivate anyone to 'get it right' and wastes trillions of dollars.

      Or the Patriot Defense System, which routinely targeted friendly aircraft during development, failed miserably the first time it was put into use(for a use it was never intended- it's never been used for what it was originally designed for, shooting down planes) and then 10+ years later was used again and resulted in the deaths of dozens of UK soldiers because it couldn't tell the difference between a helicopter traveling at less than 100 kt and an enemy missile traveling over the speed of sound?

      Or the Osprey tiltrotor, which suffered an astronomical failure rate and again, caused dozens of deaths of US marines?

      Then there's the Comanche helicopter, which they've been kicking around for years and finally decided, after spending billions, to just say "oh well, so much for that"?

      The defense department is famous for bidding scandals(if contracts are put out to bid at all), and being happy to look the other way and fudge the requirements(or ignore them completely) if the system fails to meet original requirements.

      Curiously, the russians never quite had such problems. Their fighter jets, for example, don't require pristine runways and constant maintenance; they're built like tanks, because the people who designed them knew they'd be held responsible if it failed unreasonably...and responsible doesn't mean "loose their job", it means "end up in Siberia" or "in a river with a bullet through your brain".

      This country needs three things. First, a true capitalist system for defense contractors. You want to sell the Army a tank? Fine. You can do so all on your own, without a single fucking dime, and then try and sell it. If it can't compete, too bad, your company goes under- that's the way capitalism works. Second, defense contractors need to be held responsible for when their products fail. Refunds for starters, contracts that can be invalidated on failure, civil/criminal punishments for gross design/construction failures. Third, absolutely, positively, no secret budgets of any kind. I am entirely pissed off with the pentagon filling up with all the kids who had secret treehouse clubs when they were kids and want to do the same shit now that they're 40.

      • Re:Business as usual (Score:5, Interesting)

        by LordHunter317 ( 90225 ) <askutt@NosPAM.gmail.com> on Thursday March 11, 2004 @06:42PM (#8537780)

        Or the Patriot Defense System, which routinely targeted friendly aircraft during development, failed miserably the first time it was put into use(for a use it was never intended- it's never been used for what it was originally designed for, shooting down planes) and then 10+ years later was used again and resulted in the deaths of dozens of UK soldiers because it couldn't tell the difference between a helicopter traveling at less than 100 kt and an enemy missile traveling over the speed of sound?
        Don't blame the technology when its being used in perverted ways. You yourself said that is meant for shooting down planes. It should not have shocked anyone when they tried to use it for something else and it didn't work.

        The defense department is famous for bidding scandals(if contracts are put out to bid at all), and being happy to look the other way and fudge the requirements(or ignore them completely) if the system fails to meet original requirements.
        I'd like you to name a bidding scandal then. Also, requirements are usually dropped because they were pointless in teh first place or just plain wrong. Valid requirements are rarely relaxed. Remember, requirements documents are written by committe. What sounds good on paper frequently doesn't work in real life. Anyone who's spent even a day on a goverment contract knows this.


        This country needs three things. First, a true capitalist system for defense contractors. You want to sell the Army a tank? Fine. You can do so all on your own, without a single fucking dime, and then try and sell it. If it can't compete, too bad, your company goes under- that's the way capitalism works.
        That's completely impractical. It costs too much to design a tank -- only about 3 or 4 companies in the United State could do it. Furthermore, the gov't doesn't want your tank, they want their tank. Most contracts work like this:

        • The goverment decides they want something
        • They hire someone to design it for them
        • They then pay someone to make it

        Its done this way on purpose, because the goverment likes to be in control.

        Second, defense contractors need to be held responsible for when their products fail. Refunds for starters, contracts that can be invalidated on failure, civil/criminal punishments for gross design/construction failures. Actually, they are held liable. There is this long whole process called testing, the contractor is liable until the item passes the tests. The gov't won't assume liability until it passes tests.

        Third, absolutely, positively, no secret budgets of any kind. I am entirely pissed off with the pentagon filling up with all the kids who had secret treehouse clubs when they were kids and want to do the same shit now that they're 40.
        The fact that you bring this up at all proves that you have no idea WTF you are talking about. People outside the defence community rarely understand the need for such paranoia or why we have it. But let me put it to you this way: how many security leaks do we have and have had in this country? The answer: not many. The reason: because the gov't takes security seriously, and understands it better to secure too much than secure too little.

        • But let me put it to you this way: how many security leaks do we have and have had in this country? The answer: not many. The reason: because the gov't takes security seriously, and understands it better to secure too much than secure too little.

          Actually we had some pretty bad security problems during the cold war. Does the name Robert Hansen ring a bell? But that just proves your point that security is a real threat and needs to be treated as such.

          Still, when people in the security world bungle somet

        • I'd like you to name a bidding scandal then.

          Dick Cheney's Haliburton was awarded [corpwatch.org] contracts for Iraqi oil-field work without even bidding [ombwatch.org].

          Dick Cheney's 2001 financial disclosure statement, states that the Halliburton is paying him a "deferred compensation" of up to $1million a year following his resignation as chief executive in 2000. -- from CorpWatch [corpwatch.org]

          Nope, no scandal here...

    • The problem isn't that the teams are all substandard, the problem is that (as documented in earlier Slashdot articles) the original qualification standards were set unrealistically high. Remember when everyone was complaining about all the extra hoops the teams have to jump through in order to qualify, making it seem like only a large company or university sponsored team could qualify? Well, all those hoops produced this result and they finally rethought it.

      I don't agree with making the challenge significa
    • I'd like to see a collection of teams provide their best effort, even if it falls a little short of the mark. The purpose of this is to inspire innovation, and to whatever degree it develops it's worth it. The teams that fail might have useful things to contribute, and the more exposure everyone gets, the bigger the pool of innovation we'll have to draw from.

      Remember the first season of BattleBots? Few of them would probably make it to what today's challenges are, but they threw out the ideas that ultim

  • Frank Dellaert was right (Score:5, Interesting)

    by ayatollah jones ( 692471 ) on Thursday March 11, 2004 @05:06PM (#8536882)
    From the Previous Story [slashdot.org]

    Quoting Frank Dellaert, co-director of Georgia Tech's robotics lab from the article, 'I would have trouble driving some of these roads myself. I think it's beyond the capabilities of autonomous vehicles today.'

    I guess he was right after all...

  • how fun would it be to watch a 1 team race? (Score:5, Insightful)

    by Anonymous Coward on Thursday March 11, 2004 @05:07PM (#8536893)
    If I was the one team racing unless there was a time restriction I'd go Mars rover slow.

    No sense taking risks when there are no competition.

    • Re:how fun would it be to watch a 1 team race? (Score:4, Insightful)

      by toltas ( 466545 ) on Thursday March 11, 2004 @05:15PM (#8536981)
      I think this was the whole idea, DARPA wants to make sure they have some kind of "race" on their hands, not just one team putting along till the end.

      The Carnegie Mellon entry looks to be functioning properly at low speeds, but has had some problems [redteamracing.org] staying up at higher speeds.

      I think at least having more teams will force the CM team to actually make this a race instead of a walk.
      • I think this was the whole idea, DARPA wants to make sure they have some kind of "race" on their hands, not just one team putting along till the end.

        Keep in mind that they're doing sequential starts. So it won't really be a race, just a long course with a few vehicles strung out one behind another. Far behind. And lots of disabled vehicles which have been shut down and pushed off the course (those few which even made it out of the starting gate).

        DARPA is setting up a live update page [grandchallenge.org] where you'll be a

  • Not all cases are as clear-cut... (Score:5, Informative)

    by kevinatilusa ( 620125 ) on Thursday March 11, 2004 @05:08PM (#8536905)
    ...as the article makes them out to be.

    All the Wired article states about the Caltech and Ohio State teams is that "The squads from Caltech and Ohio State University were also allowed in, even though their drones did not complete the obstacle course. "

    From the Caltech team site: "Bob completed the test route flawlessly until the last few feet. He was stopped by DARPA officials seven feet away from the final obstacle -- although had he been allowed to continue, he may have stopped himself in time..."

    Seems close enough to me.

    • Re:Not all cases are as clear-cut... (Score:5, Informative)

      by SiliconEntity ( 448450 ) on Thursday March 11, 2004 @06:07PM (#8537495)
      The Wired article was written on Tuesday night; the Caltech and Ohio State teams went ahead and qualified on Wednesday. At the time Wired was writing, it was correct that only one team had qualified. As of last night, three teams have done so. See the DARPA media page [darpa.mil] for updates on who has made it so far.

      The Wired article also speculates that even teams which don't complete the qualifier will be allowed to try the race, but I haven't seen any confirmation of that on the DARPA site. Of course, if a robot can't make it through a one mile practice track, it's unlikely to get across 150 miles of desert. But letting them try would make for a more exciting race day.
  • . . . are the same guys who set the standards for testing strategic anti-missile systems.

    "OK, I suppose it's reasonable for enemy missiles to have florescent 'HIT ME' signs with blinking red bull's-eyes and a GPS system transmitting their coordinates."

    (Seriously, the race is still pretty cool. I'm rooting for the CMU team, who used to test their vehicles in Shenley Park.)

    Stefan

  • See, this proves the Challenge was a good idea! (Score:5, Insightful)

    by metrazol ( 142037 ) <jwm33@noSpAm.cornell.edu> on Thursday March 11, 2004 @05:13PM (#8536966)
    DARPA put up a $1,000,000 cold hard cash for an autonomous vehicle and they got dozens of shoddy immitations that couldn't even navigate the test course.

    Normally they have to pay a defense contractor BILLIONS to get something that doesn't work.

    They saved loads of money, and they don't have to pay until it works, unlike, oh, other DoD projects, like the Osprey, Comanche, Patriot, TW Missile Defense, etc.
    • afaik, the comanche worked fine, but it's one of those things you don't need when there's no soviet union to fight with.
      • The Comanche seems to have successfully filled all design goals but the design goals A> did not accurately reflect reality and B> reality has become even worse for helis since the design goals were written.

        Oh well, at least we got some great video games out of it (Comanche: Maximum Overkill, anyone?)

  • You know what they say... (Score:5, Insightful)

    by Smidge204 ( 605297 ) on Thursday March 11, 2004 @05:14PM (#8536974) Journal
    When at first you don't succeed - lower your standards!

    I'm sure if every project follows this model of diminishing goals, morale will be at all time high and productivity will skyrocket! I mean, failure is a terrible thing and nobody should be forced to cope with it and try to do better. Trying is hard!

    Hey, it works for the public education system, right?

    ...right?

    Screw it. If one team qualifies, one team takes the challenge. Chances are they're a shoe-in anyway considering they've already proven themselves more capable.

    To all those that failed: Better luck next year, guys!
    =Smidge=

    • According to this [darpa.mil] several other teams did qualify yesterday.

      From the press release:
      The results of the attempts of today's group break down as follows:
      SciAutonics II, Team Cal Tech and Virginia Tech completed the course.
      Team CIMAR , Team ENSCO, TerraMax nearly completed the course.
      Axion Racing, Digital Auto Drive, The Golem Group, Palos Verdes High School, Team CajunBot, TerraHawk partially completed the course.
      The Blue Team, Rover Systems, SciAutonics I and Team Phantasm terminated their attempt

    • Ummm...you forget, maybe one team made it across the line, but if your looking for inovation, that doesn't always come from the winner. Sure the one team who had a qualifying vehicle, also had millions spent on the vehicle, where as, another team may have lots of good ideas on their vehicle and may have simply lacked the funds to develop. But having the idea is the hard part, paying a hundred engineers to design and perfect it, that's the easy part, and DARPA is the perfect candidate for such a thing.

      And

      • I think you missed my point.

        Unless I've been misinformed, the goal of this competition is to develop an autonomous vehicle.

        To facilitate this, DARPA set certain requirements that the vehicle has to meet. Consider the qualification as the first of two hurdles (the second one being the actual challange)

        Now, this isn't Battlebots. The teams are not competing directly with each other except for a piece of cheese (prize money) and bragging rights. You don't need more than one contestant at a time. It's like
        • The teams are competing by entering the qualifying runs. However, the Darpa and the teams would learn a lot more if they were allowed into the "big race." A one mile obstacle cources is a man-made tightly controlled and densly packed stretch of problems. A 150 mile stretch of land between Death Valley and Las Vegas is a much more natural laboratory, ripe with other foreseen and unforeseen problems. For example, vermin. Allowing the cars to run will provide much valuable data, even if it is to confirm f
    • Screw it. If one team qualifies, one team takes the challenge. Chances are they're a shoe-in anyway considering they've already proven themselves more capable.

      I couldn't agree more. I was in a team from my old University at an IEEE contest. Our robot had to find an IR signal that was modulated at 100 Khz with 50 mA (don't quote me on the exact numbers) through the IR emitter. Then we had to drive our robot to the emitter and do a few things, all autonomous.

      We found out that it was really challenging

    • "Hey, it works for the public education system, right?"

      Do you know anything about education? What makes you think that public education is lowering its standards? What evidence do you have for this? What's the literacy rate in the us again?

  • Lowered Expectations (Score:4, Insightful)

    by IAmTheDave ( 746256 ) * <basenamedave-sd@ ... m minus math_god> on Thursday March 11, 2004 @05:15PM (#8536978) Homepage Journal
    Well, I can't say this is a good thing. If only one qualified, then the rest need to go back to the drawing board.

    It's the same in any situation. If you lower your expectations, you'll get a lesser product/whatever. If it can't make it 200 miles, then it isn't worthy of being in the race.

    My 2 cents, anywho...

  • Funny [Ironic] Stab at DARPA (Score:5, Insightful)

    by auburnate ( 755235 ) on Thursday March 11, 2004 @05:18PM (#8537022)
    I think its funny [ironic] that DARPA added chlorine to the talent pool initially, thus eliminating many would be contestants, and now they are faced with so few "eligible" entrants. Maybe they should go back, apologize to some of the "out of the home garage" participants, and invite them to replace the a few of the multi-million dollar corporate-sponsored failures. All this from a humble robot loving gEEk.

    -- You can't spell geek without a EE.


    • 3. The participating teams are not required to develop an Emergency-Stop system. DARPA will provide and operate the E-stop for all teams.

      I like that rule, personally. Here's a quick translation.


      3. Your vehicle is not required to have an Emergency-Stop system. We have tanks which will provide and operate the E-stop at all times. Hope that allieviates your concerns!!!111

  • 4 Teams are now qualified (Score:4, Informative)

    by Anonymous Coward on Thursday March 11, 2004 @05:25PM (#8537111)
    Darpa put out a press release yesterday after
    the Wired article. Three more teams have qualified:

    SciAutonics II, Team Cal Tech, and Virginia Tech

    See: http://www.darpa.mil/grandchallenge/media_news.htm

  • I think this article is out of date (Score:4, Informative)

    by Anonymous Coward on Thursday March 11, 2004 @05:26PM (#8537120)
    Notice its dated 10 AM yesterday.

    The Virginia Tech team at least claims on its website that they completed the qualifying course.

    http://www.me.vt.edu/grandchallenge/

  • More than one team has passed the QID (Score:5, Informative)

    by Pepsiman ( 89597 ) on Thursday March 11, 2004 @05:28PM (#8537141)
    From qid_results3.pdf [darpa.mil]:

    The results of the attempts of today's group break down as follows:

    • SciAutonics II, Team Cal Tech and Virginia Tech completed the course.
    • Team CIMAR , Team ENSCO, TerraMax nearly completed the course.
    • Axion Racing, Digital Auto Drive, The Golem Group, Palos Verdes High School, Team CajunBot, TerraHawk partially completed the course.
    • The Blue Team, Rover Systems, SciAutonics I and Team Phantasm terminated their attempts.

  • autonomous robots? DARPA? STOP! (Score:3, Insightful)

    by thomas_klopf ( 672359 ) on Thursday March 11, 2004 @05:28PM (#8537152)

    Okay, besides the endless references to Terminator I could make, this really sucks. So, we have high-school kids doing weapons research free of charge for the DOD now?

    I sincerely wish that people would put more ethical concerns regarding science in the right place. While people are bemoaning the evils of stem cell research, we're happily spending money on this sort of thing.

    The happy-go-lucky attitude of the article, the competition, and not to mention slashdot is a little disturbing as well. Heyhey! That's right kids, it's time for the Darpa Competition! Just build us a robot that can run around by itself, and we'll take care of putting a gun on it. It's science-fun, just like Mr. Wizard! Even Dad's helping out!

    This isn't just "neat" stuff - this is stuff that has an impact in the real world, and I suggest that those people involved consider what sorts of contributions they're making. Personally, I would appreciate them not building autonomous robots for the DOD.

    • Apart from the military applications, autonomous vehicle development is likely to have spill-over into other areas.
      For example, reliable and cheap guidance and collision detection, consumer-level autonomous transport, hazardous environment traversal (including other planets) etc.

      Just because the military is spending money on it doesn't mean no one else will benefit. It's not like these kids are building new types of weapons.

    • I wish there was a -1: Luddite.
    • You haven't made much of a case, really. I don't really think it's ethically and morally wrong to perform research like this. I don't understand your problem with it.

      Sure, it'll be used for weapons. It'll also be used by NASA and the private sector. Why is it better to send a human into battle than a robot? I really don't feel that it's better to have an imperfect human instead of an imperfect robot.

      Sure, it's research into how to make war more effectively. But I think we can accept that war does happen,
    • Let's hop in the WAY BACK machine, and put a different spin on this...

      Okay, besides the endless references to Terminator I could make, this really sucks. So, we have high-school kids doing weapons research free of charge for the DOD now?

      I sincerely wish that people would put more ethical concerns regarding science in the right place. While people are bemoaning the evils of stem cell research, we're happily spending money on this sort of thing.

      The happy-go-lucky attitude of the article, the competi
  • they will learn to spell...

  • Let them drive through this! (Score:3, Funny)

    by Bender Unit 22 ( 216955 ) on Thursday March 11, 2004 @05:43PM (#8537284) Journal
    Ok, let them drive through this [funnyheck.com]. I mean, I just can't figure it out. :D
    (btw no need to click on the images etc. just crap below)

    • Personally I think it's beautiful. Perhaps too efficient for mere humans to navigate, but that doesn't make it bad. Can you think of a better way of interchanging between five two-way roads that takes as little time and space? (Yes, England could be blamed for having such archaic road layout that five roads could end up converging at all, but that's a different issue.) The fact that its officially called "The Magic Roundabout" suggests that the people who made it at least acknowledge its weirdness. :)

      Yes,

  • Very bad robots (Score:5, Interesting)

    by Animats ( 122034 ) on Thursday March 11, 2004 @05:49PM (#8537332) Homepage
    The Ohio State monster truck rammed a mini-van (picture) [overbot.com] on Tuesday. On Wednesday, it was stopped before running down a course obstacle. And DARPA is letting them attempt the actual event?

    The QID was pathetic. We spent two days watching vehicles move around at 1MPH and hit big, obvious obstacles. No way can most of those vehicles operate effectively offroad.

    The big design mistakes seem to be these:

    • Using a laser rangefinder aimed horizontally forward as primary obstacle detection. That doesn't work reliably on either dark or smooth objects. The black mini-van was both.
    • Using fixed line scanners. If you miss a data point, you're stuck. There's no way to take a second look.
    • Overreliance on vision. Computer vision in unstructured situations has a very poor track record.

    Only CMU is doing well. It's not the money, by the way. Their actual cash outlays are only about $300K to date. It's the body count and the fear. They have about fifty people on the project, a slavedriver boss, and the full backing of CMU. CMU has to do well; most of the Robotics Institute funding over the last three decades is from DARPA, and DARPA can turn that money off at any time.

    John Nagle
    Team Overbot

    • Only CMU is doing well.

      I wonder if the CMU vehicle really steered itself. Maybe their vaunted mapping team just scouted out the track on Monday, noted the locations of the obstacles, and designed a path for the vehicle that would go around everything. Could that have happened? You should ask them.
      • I wonder if the CMU vehicle really steered itself.

        I suspect that it did. They have the technology. But I wonder about some of the entries that totally blew it on Tuesday, then did much better on Wednesday.

        The original QID plans included a moving car-sized obstacle to be avoided, but no such obstacle was present at the real QID.

    • Was that a flaw? (Score:4, Interesting)

      by SuperKendall ( 25149 ) * on Thursday March 11, 2004 @07:02PM (#8537935)
      From the look of things, the Ohio State Robot just decided that it could go right through something as tiny as a van - I think it had the right idea, they just stopped it before it could uttery crush the van.

      Was there some kind of rule against destroying objects in your path instead of navigating around them?

      Actually I am half serious as what happens to an automated supply convoy when the lead vehicle is destroyed by a mine? You'd hope the remaining vehichles could just push the thing out of the way and go on.

      • Was there some kind of rule against destroying objects in your path instead of navigating around them?

        Yes. The rules are very explicit about intentional crushing.

        • Dissapointing, but I suppose it's a good idea so one team with a good hunter-killer bot doesn't stop anyone else from winning!
    • Only CMU is doing well. It's not the money, by the way. Their actual cash outlays are only about $300K to date.

      I find this an awkward comment. How can you say it's not the money, when these guys have gotten incredible amounts of equipment donated by their sponsors? Sure, that's not cash out of their pocket, but someone is paying for it.

      Their site is one of the most commercial websites I've ever come across with almost every other line having a link to a sponsor.

      And we're not only talking hardware being
    • Given that they are letting everyone participate now without passing the qualifier do you regret not entering, just to get some race experience/data?

  • TerraMax photos (Score:3, Informative)

    by incuso ( 747340 ) <incuso AT gmail DOT com> on Thursday March 11, 2004 @05:56PM (#8537407)
    Photos of our vehicle: http://vislab.ce.unipr.it/terramax/
  • ... then they shouldn't drop the requirement. Just let the CMU entry go it alone.

  • Check the dates! (Score:3, Informative)

    by KenBot_314 ( 744719 ) on Thursday March 11, 2004 @06:02PM (#8537455)
    C'mon Slashdot!

    The official rules linked from the Wired article have not been updated since January 4, 2004.

    The rules actually refer to the field test part of the QID as a Demonstration of basic abilities. It never says anywhere in the rules that they have to fully complete the demonstration to be qualified.
    The rules have always stated that 25 teams would make it to QID but only 20 teams would actually compete in the race.

    Yet another shining example of the media trying to make news where there is none...

  • NPR news story on 3/13/2004 (Score:4, Informative)

    by redwoodtree ( 136298 ) on Thursday March 11, 2004 @06:19PM (#8537582)
    Here in the states NPR had a good story on one of the teams yesterday (wednesday) during their All Things Considered show. You can listen to the show here:

    NPR LINK [npr.org]

    The NPR Summary of the story is: At the crack of dawn this Saturday, a 200-mile race across the Mojave Desert begins. The competitors are robotic vehicles taking on the form of SUVs, dune buggies and golf carts. It's a contest sponsored by the Pentagon to spur advancements in the field of robotics. NPR's Melissa Block talks with competitor Red Whittaker.

  • Red Team can't really "win" in my opinion (Score:4, Interesting)

    by ChiralSoftware ( 743411 ) <info@chiralsoftware.net> on Thursday March 11, 2004 @06:24PM (#8537624) Homepage
    The Red Team is favored by everyone to win, but is it really winning? What they have done is constructed an amazingly accurate and detailed map of every last bit of topography, down to the size of a big rock, of the region the race is going to be in. Their on-board sensors and navigation equipment doesn't have to do much sensing and navigation at all; they will get a foot-by-foot map of where they should be going. This doesn't strike me as "autonomous". It strikes me as just another version of remote control. Their victory will be an impressive technical feat but it certainly isn't the same as having a vehicle that you can plop down on unknown rugged terrain (be it a war zone or the surface of Mars) and have it get around on its own.

    I will be more impressed if the autonomous motorcycle makes it ten miles than I will be for Red Team to win the whole thing, because at least this bike is fully autonomous and has some radical new ideas going into it, instead of just tons of resources and brute-force mapping.

    ----------
    Host your WAP site [chiralsoftware.net], automatically

    • Re:Red Team can't really "win" in my opinion (Score:4, Interesting)

      by feelyoda ( 622366 ) on Thursday March 11, 2004 @06:39PM (#8537763) Homepage
      Well, there is certainly is misunderstanding of what it takes to do this.

      A "negative obstacle", i.e. a hole, 1/2 meter deep could very easily be missed by the map. This would cause most cars to crash, and is very hard for even humans to detect.

      This is one example of dozens of things you MUST perceive in real time. To say that the Red Team isn't really autonomous is insane, and you have little appreciation for 1) their action setup, 2) how hard the problem is.

      Besides, humans most certainly have some sort of impressive map making capabilities that let you find the bathroom with no lights on. That is no small feat.

      Look around redteamracing.org a bit more to learn what they are actually doing before letting your jealousy get in the way of your head. (Also note that the motorcycle is ridiculous. Since when is it easier to keep a two-wheeled machine stable under off-road conditions than a hummer? They have engineered their own failing...
      • Actually, the motorcycle entry (the Blue Team, from U. C. Berkeley) is the most innovative vehicle there. Not only does it self-balance, but the computer hardware is quite impressive, with FPGAs used for vision.

        It's not going to win, and they don't expect it to. It's a technology demonstration. I've met the people doing that project, and they have a coherent vision. They view the Grand Challenge as a beginning, not an end.

    • Can you think of an environment where an autonomous vehicle would need to be on the ground before it could be mapped by a flying autonomous vehicle? As long as the US has space dominance, anything the DoD is paying for will have satellite photos before it sets down on the ground. And Mars is being carefully mapped as we speak...

      It's just good design to divide an autonomous system into a seeing part and a moving part -- you can't see as well from the ground no matter how good your optics are.

    • This is not necessarily a bad thing. Mistakes in problem specification lead to unwanted results, but also lead to more accurate problem specification.

      There was an AI team in the UK that wasa trying to make a box with wings fly. They gave it some rudimentary wing movement instructions and a genetic algorithm with the fitness defines as distance from the ground.

      Instead of flying, it propped itself up on its wings. Thinking about going sround the problem is sometimes as good as going over the problem.

  • Negative Spin (Score:4, Insightful)

    by SiliconEntity ( 448450 ) on Thursday March 11, 2004 @06:38PM (#8537747)
    I'm concerned that the spin on this event from much of the news media is negative. It's starting to look like DARPA will end up with egg on their face if none of the vehicles do well.

    What these writers forget is that the event was intentionally designed to be incredibly difficult. In earlier news releases, the idea was expressed that this would be something that would be run every year until a robot manages to win it. This is in the spirit of other super-difficult prize competitions, like the X-Prize or the ancient quest to develop a method to compute the geographic longitude of a ship.

    It's too bad that an inventive, flexible and interesting approach by DARPA is being spun as a failure just because the first tries haven't been all that successful. I'm really hoping that no teams win and that DARPA does run it again next year, because by then we'll have many more good contestants. This year's entries will have gotten the basic bugs out of their systems and be genuinely ready to tackle the course; and there will be a few new entrants as well, finishing up at the last minute and just hoping that they have something that will run. Each year will see improvement. To me that would be far more interesting and enlightening than a one shot deal.

Slashdot Top Deals

Solutions are obvious if one only has the optical power to observe them over the horizon. -- K.A. Arsdall

Close