Grand Challenge 1, Competitors 0

Ivan writes "According to the DARPA Grand Challenge Status Board, 2 bots were withdrawn before the race started and the remaining 13 were all disabled. Red Team and SciAutonics II tied at 7 miles, a bit short of the 142 miles required." CNN has coverage and interviews.

Comments

Is there still a chance....

[Dr Reducto]

Has the "privateer" race been done? It would be nice to see a privateer complete the challenge.

I think that even though they only got 7 miles into the course, thats still damn good engineering. Maybe next year they will have worked out what has gone wrong and figured out a way to flesh out an autonomous robot (Or hide a midget navigator somewhere!).

Processing power

[Ephboy]

The whole thing makes you conisder just how much processing power we use to control our speed around curves and avoid potholes when we're driving. We can integrate a hell of a lot of information, process the relavent signals and adjust our behavior in milliseconds. And that's not adding the additional struggle of trying to get your iPod to play through the stereo system....

This is *great* news!

[myowntrueself]

It means that autonomous fighting machines are still some way off.

I suspect that the first industrialised nation that develops autonomous fighting machines will take over the world (or at least have a damn good go).

Re:This is *great* news!

[myowntrueself]

The 'battle bots' are not autonomous IIRC, they are remote controlled.

Autonomous fighting machines would mean that even a nation of cheese-burger munching, channel surfing couch potatos with the reaction speed of a head of broccoli could have a go at taking over the world.

You wouldn't even need to enlist l33t gamerz to pilot them by remote control and risk the communications being jammed or having remote control operators charged with war crimes when they get too entheusiastic.

They are ideal; there would be no need to take and hold populated land.

If one wants the oil or other minerals one would be able to unleash ones autonomous machines to exterminate the human population.

And when the UN says 'but this is a war crime!' one merely passes the blame to the manufacturers and software house which designed the systems.

They then point at the EULA which absolves them of all responsibility for anything that their creations do.

Since no human actually committed any massacres, and no human officer gave any specific orders for the machines to slaughter men women and children, no human is responsible and one can simply rid oneself of a troublesome occupied population.

Who wants future war like this?

Re:Really pathetic showing?

[Hello this is Linus]

One of the rovers got tangled with barbed wire, another had a malfunctioning satalite navagation system (GPS?), one broke its axle, and one rolled over... So its not as easy as it seems, the terrain must be pretty rough to have a car over turn or break its axle.

Still, might have been better to start small

[btempleton]

Yes, this was a grand challenge. But it would be nice if teams could solve part of the problem at first, get some recognition and minor prize money for that, and then move on.

So perhaps step one should have been just doing a long ordinary road course, minimal obstical avoidance, just handling roads, turns, potholes, ramps and even traffic lights (where you are told they are).

That contest would provide useful civilian tech and also useful military tech in terms of a autonomous vehicles to carry cargo in a controlled area with intact roads.

Or you could also imagine autonomous vehicles which handle roads, but then get to a rough patch they can't handle. At the rough patches you station soldiers who drive/remote control the vehicles over the rough patch, but you need far fewer because they stay in one place and only do the rough patch. Let humans do what they can do and computers do the boring long-haul road drive.

Next, hold a contest for a shorter rough course with obstacles.

Finally, combine the two.

Re:Still, might have been better to start small

[Mr_KnowItAll]

Having just now returned from the starting line, I spent the drive home thinking about how the next GC should be changed to make it more valuable (relative to the stated objectives). Rather than shortening the course or creating more obstacles, DARPA really must refrain from giving the human team the route before the robot is released. Today's teams had two hours to review and perform detailed planning on the race route to "pre-program" their vehicle rather than force it to rely on on-board intelligence. This is entirely useless for a working robot. It also provides an unfair advantage to a team that has manpower and resources to scour the possible courses in advance. The next DGC could be made much shorter and more difficult by making it a true test of intelligent, autonomous navigation over a route that is given only to the robot.

The value of stupid solutions

[pdxdada]

What really surprised me about this competition is that no one went for the stupid solution. I read about a robot competition in the late 90's that involved obstical avoidance. One of the top finishers had no computer controller it only changed course when it ran into something and placed high just because it didn't break down.

Lessons?

[Quixote]

I submitted the following as a story for some discussion, but it got rejected; so I'll just post this for discussion.

Just like 1000s of geeks worldwide, I watched this with great interest. But the whole organization thing left me with a funny taste in the mouth.

It seemed as if the group that could throw the most money at the "problem" would win. Take the CMU team for example: they paid for a high-res survey of the area; had undergrads map out each and every obstacle in all of the possible paths; etc. Now, if the goal of this "grand challenge" was to unleash the entrepreneurial spirit, then it failed. Money != Entrepreneurial spirit.

Taking lessons from the RoboCup people, I would have preferred that DARPA organize it as follows:

1. Create a simulator for the sensors, and design a small (virtual) course for this simulator. Let people develop algorithms using this simulator, and have a competition in this virtual simulator to select a set of (say) 30 teams.
2. Provide each of these teams a platform: a humvee equipped with the sensors (actual ones from (1) above). Of course, if a team wants, it can add its own sensors.
3. After some time, hold a "grand challenge".
4. Analyse the approaches taken by the various teams, and (important) share the code among the teams. If a team designs a new sensor that is useful, get copies made and share with the teams for the next iteration.
5. Go back to step (1) above, and repeat.

Preference could be given to schools or efforts involving students, as not only is this a great learning experience, but also it will be a great motivator.

Just look at the technology gap between CMU and the rest of the entrants. It is quite an achievement that someone was able to equal CMU in performance.

There are a lot of smart hackers out there who would love to take a crack at this problem, but the lack of hardware is a serious hurdle.

Re:Lessons?

[Quixote]

I think it would be safe to say that the mechanical (platform) aspects of this competition are insignificant compared to the sensing and control aspects. With a human driver, most of the entrants would have completed the course with ease.

In my suggestion, by providing interested entrants with a common platform, you take away these secondary issues, and get to focus on the primary issue: how to use the sensory data to control the vehicle so that it can get from Pt A to Pt B.

Re:Lessons?

[Jodka]

Your suggestions show that you do not understand the purpose of the contest. That purpose is to compare different designs and methods of problem solving. To the degree that DARPA constrains those designs and methods with regulations, as you are proposing, it suppresses diversity and experimentation, destroying the value of the contest itself. To be specific:

1. You don't encourage people to think outside of the box by sponsoring free boxes. The same goes for providing humvees. If you limit everyone to start with the same platform then you won't get people experimenting with different platforms, which is desirable. We saw extreme variety in this matchup, from a motorcycle to that massive truck from OSU. This diversity is good because you learn from it which designs work and which do not.

2. The participants are not just developing robots, they are also developing methods for developing robots. If DARPA constrains participants to all use the same development track, then they are undermining a significant experimental aspect of the contest.

3. That different teams have different budgets is a good thing, not a bad thing. Part of what you want to learn from this is how much performance do you get per dollar. The only way to learn that in the real world is to let different teams spend different amounts and correlate performance with cost. The military has a huge problem with this tradeoff already, and I expect that's one of the questions that they want this contest to help answer. There is increased skepticism about the conventional military practice of purchasing the near-perfect weapon at near-infinite cost. NASA adopted the "faster, cheaper, smaller" agenda to get away from that kind of spending. But where is the sweet spot ? There is a legitimate question here about what is the optimal number of eggs to put in one basket. That's something you learn by letting different groups spend different amounts to solve the same problem.

I mean look, if it turns out or not that the only way to solve the navigation problem is indeed to have students map obstacles in advance, then DARPA has learned something by allowing that expensive strategy into the contest. CMU has more money so they can try that approach. Someone whith less money might experiment with something more innovative. These robots are both spending experiments and technology experiments. DARPA does not want every contestant to use CMU's expensive strategy, because that gives no comparison case. Uniformity bad. Diversity good.

4. The entire reason to encourage development in this area is that the military does not know how to develop these robots. If it did, it would not need to hold the contest. So why dictate to contestents a procedure for developing robots ? Different groups will use different methods. Some will fail and some will succeed. You learn from that what are the better methods.

5. The great thing about the absense of such regulations such as you propose is that people like yourself who are convinced that they have the best rules for how to develop a robot can try their ideas, or promote those ideas to actual contestants. You don't need DARPA rules to dictate your own strategy to yourself. If you think that your scheme for how to develop a robot would win the challenge, then why aren't you using it yourself, or trying to convince a particular team to take advantage of it to win the prize ? That instead you want DARPA to force your favorite development methodology on all contestants suggests that you have low faith in your own ideas. If your ideas are so good, why do the rule makers need to force them on people ? We should all be suspiciuous of arguments such as yours, those of the form "My plan is better for everyone, therefore you should all be forced to follow it". When someone says that, what they usually mean is that their plan is so bad that the only way anyone would follow it is if they were forced to.

6. Not just you, but a lot of other people are convince

The words ring true

[GillBates0]

From my earlier accepted story:

DARPA Grand Challenge Kicks Off March 13th
Monday March 08, @10:40PM

GillBates0 writes "A quick reminder that the DARPA Grand Challenge is due to kick off March 13, the coming Saturday." He points to this "quick recap of the teams participating in the event," as well as details about the available satellite feeds. "The Atlanta-Journal Constitution is running a story about the event today. 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.' (shameless school plug). We'll see if the participants can prove him wrong."

Those words ring so true now...I never expected the contest to end on such a negative note.

try this at home

[Kappelmeister]

I was thinking about that a few months ago when I was working on a film shoot. I had to run some supplies between a remote location and the base camp. The terrain was really rough-- this was a forest, and it had a lot of trees, roots, rocks, crevices and the like. And I absolutely, positively had to be at the base camp ASAP.

I found myself running as fast as I could, but my mind set up an interesting pipeline. I was always looking five to eight feet in front of me and my brain feverishly tried to parse out what was a rock, what was a branch, what was a big root, what was sloped ground, what was even ground, etc. Then, that information got passed to a route-choosing part of my brain that decided where the best place was to put each foot: left, right, left, right. That information, in turn, went to my brain's motor department, which was actually in charge of balance, weight distribution, and muscle movement to actually put the feet where they were supposed to be and keep my momentum without falling.

I call this a pipeline because my eyes never left that five-to-eight foot range. When I was selecting any bit of route, I was already looking at the next bit of route and stepping on the last bit of route. I never looked at my feet, but somehow always put them where they needed to be.

I wouldn't make such an analogy anywhere other than slashdot, but I could feel that the load average on my brain was as high as it could be. I didn't have any free cycles to think about my day, or have a song in my head, or think of my next joke, as I usually do. Every ounce of my concentration was going to these automatic, practically sub-conscious processes. I know was processing as fast as I could -- any faster, and my brain would tell me, "I can't parse the terrain that fast," or "I can't decide on a route that quick."

Don't give me any credit for it, because it has nothing to do with knowledge or intelligence, but I was solving an extraordinarly tough problem very quickly. In short, if I could bring my brain to the edge, I can see how tough this is for the DARPA contestants!

I'm surprised no one finished but...

[Garak]

I'm really surprised no one finished the race but from experience I know how tough robotics can be.

A few years ago I was a member of a Highschool robotics team were we build a hockey playing robot one year and a tank the next. They were RC which made is doable but it still was quite challanging.

Robots don't have self preservation instinct and usually end up destroying it self. This seems to be the case in this competition.

When driving a car your not only trying to navigate and not hit people, other cars,etc... Your also trying to not beatup your car. This is a very hard thing to program into a robot. Driving up a rocky hill isn't a simple as taking path with the least rocks in the way, usually its better to find another way around. But in programming how the hell do you tell that its unpassable. A brick wall is easy but a washed out road is hard to determin with cameras and other sensors.

As a human one would get out of their car and walk through before driving. In a race situation you would already know the course and whats passable.

Another challange is sand, its very easy to get stuck and its also hard to tell how deep or lose it is.

In miltary applications you would have detailed aerial photos or beable to take your time so this isn't a realistic test.

Re:Really pathetic showing?

[EvilTwinSkippy]

You have never been to the Mojave Desert. It's not 200 miles of beach. There are cliffs, dunes, meth labs in trailers, unexploded ordinance from military training exercises, and so on.

What did you think dune buggie were designed to handle?

The trouble spot

[Animats]

Both CMU and Caltech seem to have failed on Power Line Road near Camp Rock Road, That's a rather boring piece of terrain, and seemingly easier than the first two miles. Again, what went wrong?

what can we learn?

[snarkh]

Absolutely right.

In fact, it is not even clear what we can learn from failure like that, which we could not learn otherwise.

Flashy things like this race do not necessarily tell us anything more about deep problems of AI. One can spend millions and millions and not get any closer to the goal.

Re:Rough terrain's a bitch

[primus_sucks]

Kind of make me wonder if a biped design might be the way to go.

Pulling a team together

[Valkyre]

I'm looking for sponsorship, fabricators, and programmers in the MN/WI area to compete in the next competition. Interested parties please contact darpachallenge(at)phayze.com Serious offers only. Must be able to put in 20+ hours/week minimum and work with a team.

Why not jump instead of roll?

[NewtonsLaw]

Given that the terrain may be filled with obstacles and uneven in contour, why didn't someone come up with a machine that simply jumped 30-40 yards at a time -- using a combustion ram to provide the motive force.

This would have potentially overcome many of the problems and if it were designed to be self-righting, even landing on an award contour and rolling over shouldn't be too much of a problem.

Another alternative is something that had short-term "hover" capabilities -- ie: checked the path ahead was clear for the next x-yards and then, drove that distance. If it saw something that appeared to be an obstacle it could hover over it for whatever distance was required.

Come to think of it -- why were DARPA so all-keen on using wheeled vehicles? What would be wrong with a hovercraft -- even one without a skirt so that the barbed-wire wasn't so much of an issue?

Wheels are okay, but they're certainly not the best option for uneven and unpredictable terrain -- after all, nature is an *expert* designer but you don't see any animals with wheels do you :-)

Re:Why not jump instead of roll?

[Xepherys2]

Well, jumping isn't a great things, as it would make the vehicle that much more obvious/visible. DARPA isn't trying to have something invented that looks cool or "just works", it has to serve a purpose.

Hover vehicles have limitations of their own, that have actually been discussed on /. boards in the past.

As for nature... you don't see any animals that "hover" either... perhaps save the flying fish. You have some that fly, and a few that float down from somewhere higher up, but not more than 1 or 2 that have any type of actual "hover" capabilities.

Wheeled craft are easier to maintain than jumping or hovering equipment. That means that the military that would inevitably use this technology would not need special training on the mechanics... just possibly on the electronics used. Quite possibly not even on that.

Personally, I don't get why they didn't use low, bumper-mounted radar to detect things like giant obtrusions so that axles didn't get broke and the like. *appauled* I really don't get it!

Re:Sharing

[Quixote]

No, you share after the competition.

An analogy would be genetic algorithms. Once you evaluate the fitness, you perform crossover. By making the code public, teams are free to pick up the best ideas from the others, and build on each others' work.

Take a look at how RoboCup soccer works. They have made great strides in the last couple of years, and a lot of that is due to the spirit of cooperation in the setup.

I wouldnt say that.

[rebelcool]

in some autonomous robotics stuff i've worked on, after task competitions teams would get together and discuss varying approaches to problems.

While no code sharing occured (nor would've been useful, as each platform has its own unique way of doing basic tasks), discussing approaches leads often to combined approaches, fresh perspectives on ideas and then THAT leads to innovation as each time takes what they've learned and applies it to the next project. Eventually the most efficient and "best" system results.

By your logic, nobody should go to a university to learn things and should learn everything on their own to foster 'innovation', while in fact everyone would be reinventing the wheel a billion times over.

Re:TROLL EXPOSED: COPIED FROM ALASKAN ENTRY

[aardvarkjoe]

Not only that, this guy has posted quite a few posts today -- all plagiarized. (Google turned up a match for every single one I looked at.) Since I happened to have mod points, and I usually don't use them, I went and moderated a bunch of his posts as troll. Anybody else want to get in on the action?

New world record?

[LostCluster]

Seeing this was the first-of-its kind event, is the last bot out now a world recordholder for the distance it traveled? Maybe the 150ish miles goal was a bit too much to ask of the new technology, a bot that can travel 7 miles reliably has some use, just not as much as DARPA was hoping for.

Also, was there a particular barrier at the 7 mile point that did in the last two robots, or was it just coinsidence that they both stumbled within a mile of each other?

Re:Congratualations to those that tried.

[tftp]

On a battlefield, sometimes even half a mile is suicidal. If you have your men cut off from supplies by such a stretch of, say, bombarded road, or snipers, what do you do?

In World War II a soldier was sent to drive a truck, and if he fails... then another, and another. Today we can send such a robot. It is safe from snipers, and if it gets hit with a shell it will be simply replaced.

Machines like these can -already- be used to patrol large territories; with improvement, they will be really good at that.

On Winning

[macmurph]

You have to engineer the process of winning, not just the technology to win.

I worked on a solar powered race car that was to cross the country. Our superior car won the first few days, but eventually crashed.

I learned a lot more about team work and egos than I did about technology. The technology was there, the money was there, the open-minded cooperation was not there. The car was engineered very well, the win was not engineered at all.