Slashdot Mirror
DARPA Announces Grand Challenge 2005
An anonymous reader writes "The Grand Challenge 2005 Date has been announced for October 8, 2005. Check out DARPA's official webpage for details. Already several teams from last year are gearing up: Carnegie Mellon Red Team, D.A.D., and Cal Tech. Also, several new teams are entering, among them Stanford, and Florida Tech. Should be a very interesting Challenge next year!"
I've actually seen something very similar to this before, I believe that they should be able to actually do this. At a local college near me I watched as the robotics team ran their robot through an obstacle course by having it follow a nice little line using an optical sensor... So I think this challenge will be possible, although not easy to do for one mile. But even if noone actually does claim the prize and finish... it is the fact that you learn something from the experience and hopefully that experience can be put to good use that's important... isn't it?
"Should be a very interesting Challenge next year!" ...more so if anyone finishes
More interesting for who? The crowd or the researchers? Hey, if you want an exciting race, go watch NASCAR (or not). As far as the researchers are concerned, it's probably more interesting when these things fail to finish the race than if they all completed the course without difficulty. Each failure teachs the researchers something about AI. These "lessons learned" are then used for a variety of applications and theoretical extensions, not just building a better autonomous car.
DARPA is not putting on these contests because they expect someone to win. They are trying to give researchers a difficult problem to work on. Don't be looking at the fact that none of the vehicles came close to finishing the race as some sort of failure or "boring". The problem is quite exciting for the researchers and for anyone interested in AI. The fact that they are making it an annual thing is icing on the cake.
GMD
watch this
This topic, although w/ slightly different wording has been chewed and swallowed/spat out couple of days ago...
the same spooks that want to identify terrorists by how they walk?
No, they're the ones who created the Internet. You know, that thing you're using right now. Plus, they happen to be a huge supporter of scientific research and development. Given the fact that industry has slashed R&D in favor of raising the quarterly reports by a half a percentage point or giving the members of the Board an ivory backscratcher as an annual bonus, let's not be too quick to insult one of the few remaining patrons of science.
GMD
watch this
because it would be unfair to newcomers to the competition.
OTOH, what top 15 teams? There was a TOTAL of 15 teams qualified!!!
Make it go up to 60 MPH to make up time when it doesn't have any barriers to get around, and let it go
And what will you do when this thing crashes at 60 mph into one of the barriers you haven't designed it to detect?
Collision avoidance without physical contact is one of the biggest challenges these teams face.
Also, the obstacles aren't necessarily these big, obvious concrete barriers dumped in the middle of the desert. How would your robot deal with encountering a lake, or coming to the edge of a cliff?
This space intentionally left blank.
It should be noted that the Aerial Robotics Competition didn't have any winners for its first few years of competition. Each year they changed the rules to make it harder, even though no one finished the previous year's challenge (in fact, no team even had a vehicle capable of autonomous flight). Then, one team had a helicopter that autonomously hovered, navigated obstacles, and even located and picked up an object off the ground -- while no other team could even autonomously fly. The following year, every team had a flying entry.
The Darpa Challenge might not have a winner for a couple of years, but if they keep raising the bar a little bit, results will follow.
DARPA is not putting on these contests because they expect someone to win
thats right, they are doing it to use the concepts and ideas into devices that aid them in a war, either killing people more efficiently or saving their own troops lives , either way its getting school kids to think up better ideas to gain the advantage on the battlefield, any wonder as a nation usa is fucked up
missing the whole point, technology didnt help win the war in vietnam, or Iraq so its value on a battlefield is questionable, defense doesnt have to involve violence, maybe when the country is a bit more mature DARPA might get round to realising this, untill then keep sex off tv and teach your children war instead
DARPA is the central research and development organization for the Department of Defense (DoD). The Agency manages and directs basic and applied research and development projects for DoD, and pursues research and technology where the risk and payoff are both very high and where success may provide dramatic advances for traditional military roles and missions.(emphasis mine)
This mission statement is important because it emphasizes that DARPA was specifically designed to fund research into hard problems. In fact, the catch-phrase "DARPA-hard problem" has surfaced as a shorthand way of labeling high-risk, high-payoff problems. A lot of people seem to be having fun laughing about the fact that none of the vehicles in this year's race came close to finishing. But as I pointed out above, that certainly doesn't make this contest a failure by any stretch of the imagination. This is a DARPA-hard problem. Certainly no private company is going to fund something like this. The risk is just too high. Really, DARPA is one of the few entities that has the resources to put on an annual contest like this.
From some of the comments posted here there seems to be the idea that because DARPA is the Defense Advanced Research Projects Agency that we have to dislike them. I disagree. Sure, DARPA has future warfare in mind. But so many technologies developed for military use end up in the private sector anyhow. It would be pointless to list them all. Let's not look a gift-horse in the mouth. DARPA is willing to fund science and scientists are willing to work on their problems. I don't see what the problem is. Do you guys think that if scientists stop doing DARPA work that suddenly conflict in the world is going to stop and everyone is going to join hands and sing "We are the world"? Sometimes I think people complain just because the like to complain.
GMD
watch this
Hey, if you want an exciting race, go watch NASCAR (or not). As far as the researchers are concerned, it's probably more interesting when these things fail to finish the race than if they all completed the course without difficulty
Hey, NASCAR races are more exciting when that happens too. People love to see stuff break.
Here is google's cache of the old rules page.h iIaoMJ: www.darpa.mil/grandchallenge/rules2.htm+darpa+gran d+challenge+rules&hl=en
http://216.239.51.104/search?q=cache:QHHtt
Here are the two relevant sections:
"3.1 Team Must Be U.S. Entity
The Challenge is open only to US entities. This includes U.S. corporations, U.S. non-profit organizations, U.S. universities, U.S. citizens, sole proprietors that are U.S. citizens or permanent residents, and partnerships of U.S. citizens or permanent residents. The nationality of the team is determined by its Team Leader which may be a person or organization."
"3.4 Foreign Participation
Foreign entities may participate only as members of a U.S.-led team."
With the exception of perhaps the dirt bike, I don't htink the choice of donor vehicle made a very significant different. The Hummvee should have been pretty darn formidable, but it wasn't.
Jsut like having a Ferarri doesn't make you Schumacher, starting off with a Range Rover isn't going to make the software any better.
Another way to deal with it is with a range finder angled towards the ground- if it's suddenly WAY steeper in front of the robot than the robot was expecting, turn around and find another way.
Speaking as a former off-road enthusiast, unless that range-finder is suspended a couple hundred feet in front of your vehicle, it's not going to do you any good if you're moving at 60mph. Stopping on natural surfaces can often take a lot more distance than you're used to, especially if there's a downgrade involved.
Add to that the fact that hitting an 18" deep cut in the surface at high speed can tear the wheels right off the vehicle...
Seriously, running at 60mph on established, familiar trails in the open desert will keep your brain running at full speed trying to keep you safe. Running at anything over 40mph in unfamiliar, non-graded landscape is seriously risking your life. Most of the time I never got much above 30mph, and even that was hard on the vehicle. (Two center hanger bearings, six CV joint boots, a differential, a steering stabilizer, four shocks and an A-arm in one year. Plus numerous dents, scratches and bruises.)