Slashdot Mirror
Fifteen Teams Selected for DARPA Grand Challenge
doughnuthole writes "The official posting has been made of the 15 teams that qualified for the Grand Challenge, seven of which completed the entire QID course. The top three teams, and thus those who get to start first, were the Red Team, SciAutonics II, and Team Caltech. The race starts at 6:30 am Saturday, with teams leaving every 5 minutes. A live webcast will be available at grandchallenge.org." Reader uss_valiant writes "Tomshardware runs an article about DARPA's Grand Challenge. It features new pictures, the DARPA video of the qualification and covers some technical challenges such as the obstacle detection."
For all those of us who don't have access to the DARPA channel, we can stream the telecast live from here.
These are the same people who appeared in this slashdot story and seems to be different from the "live webcast" mentioned in the story which only appears to have a tracking feature.
If a team leaves every 5 minutes, (and assuming the first few hundred yards is relatively easy going - you find that on most courses of any nature), then we are going to have an awful lot of bunching at the first point the vehicles start dropping below 25mph. Interestingly, the rules state that the team in front (i.e. being passed) has right of way, unless E-stopped.
Exercise your right not to vote. thinkoutside.org
6:30? That means nothing in nowadays world!
WHAT TIMEZONE???
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
CNN is running a story about the DARPA race/challenge, in case anyone is curious.
And the odds on any of these machines even finishing the challange? Pretty slim, Red Team looks to have the best chances, pretty nice looking machine they've got going as well. All in the name of science and progression I guess .. but if the army vehicles auto targeting equipment couldn't distinguish the difference between a helicoptor and an incomming vehichle .. what are the odds on the software they put on the 'finished' development being any better? also pretty slim.
May as well just spend the money on deveoloping the something worthwhile.
If at first you DON'T succeed, Skydiving is NOT for YOU!!
Informative? Wait I can do this do:
Last year I had a pottery course with a person who claimed her sisters, friends, dog trainers mother in law actually handled funding at DARPA. She said that this was actually a nefarious plan to speed up the development of killer robot cars that will one day rule the world!
How is this funding a university? The prize of $1m is unlikely to or perhaps only barely just cover the costs of any serious entry. People clearly aren't in it for the money.
Read reviews of shopping cart software
The prior event was the QID it was designed to determine the saftey of the vehicles only. they didn't change the rules mid-event (as the earilier post would have you believe). But a prior to the start of the QID.
A lot of folks are spending tons of time and energy attacking a hard problem. If DARPA thinks they deserve to watch their car leave the "official" start line, it's DARPA's decision to make. It's DARPA's event, they can run it however they want to, if they wanted to tape rubber ducks on the hoods prior to departure, they could ask eveyone to do so.
Winning the race might be worth $1m up front... but how much is a contract to build robotic vehicles for the US military worth? And of course, many universities would be researching automated robots even if the competition did not exist; winning an extra $1m is just a bonus.
According to previous comments at Slashdot, a map of the course has been leaked, meaning that entrants can cheat by pre-programming a course.
Its also a lot of publicity for the university... and I'm sure many an academic paper will be written by the teams... publish or perish!
When I was a Freshman at CMU, I worked on a research programming project translating a bunch of the Navigation code for this self-driving HMMWV ("Hum vee"). It was originally written in C, and we converted it to Ada... though I bet it's been reverted back to a more versatile language.
It used multiple cameras mounted at different heights to build a 3D view of it's surroundings, and could judge all kinds of obstacles... though at the time (7 years ago) had a lot of trouble with streams and shadows. I was amazed that it could recognize stoplights correctly, and even signaled when it was changing lanes on a street.
Either way, it was a great project for a young would-be programmer to work on, very amazing stuff, and lots of cool toys to see in the Robotics Institute there.
-Hell hath no fury like that of a woman scorned for
Actually I watched a show last night and they Highlighted the Red Team's Vehicle. After watching the show I was given no idication of any other competitors, but they didn't provide any reasons why the Red Team might not finish. Hell, after what I watched I'm wondering if any other teams are really even playing the same game... That is one high priced Hummer.
Post: Sigged, for your pleasure.
Assuming that "6.30 am" means local time in Barstow, California, that would be 6.30 PST -0800, or 14.30 GMT.
Updated on the live status board A bit short of the original 250miles.
What we haven't landed on Mars yet?
Look, folks it's engineering. It takes time. Frankly, competition is good. You have to understand, most of these schools/people participating don't make multi-million dollar robots for a living. CMU is probably the best (where's MIT??). Maybe CalTech or Berkeley is a close second. We didn't win the space race overnight. Engineering takes time. Eventually, the competition will learn the best techniques and everybody profits. It's is an educational thing...
DARPA checklist:
-sentient AI
-robust hardware design
-massively parallel neural net
-robust error handling
-programmed fundamental laws of robotics
-able to withstand a tank blast
-able to withstand a bomb shell
-able to withstand a nuclear/biological/chemical attack
-able to withstand a REALLY BIG MAGNET!
Seriously, I think even Sadam could beat our robots! Just buy the mother of all big magnets (or make one). Oh that's right, they need electricity! Sorry, carry on. Maybe they could get a donkey to run on treadmill and make a generator.... (Okay, not so seriously.)
So, how robust can any robot be? All I need is a really big magnet and it's screwed.
Yeah, how come the Terminator/Matrix/Inspector Gadget never had to worry about magnets?
What do you mean my sig is repetitive? What do you mean my sig is repetitive? What do you mean....
Team Terramax is a collaboration between Oshkosh Truck and Ohio State University based on the MTVR, a six-wheeled, 425 HP, seven-ton truck.
I had the privilege of test-driving an MTVR on the obstacle course at their factory in Oshkosh, Wisconsin. This truck, fully loaded, could take on hills steeper than feel comfortable taking on in an ATV. We forded streams, climbed over barricades, and did steep side grades, all without breaking a sweat. I've got no doubt that this vehicle is up to the Grand Challenge, if the guys at OSU have their technology in order...
The cure for cancer is coming: Reovirus
Info
The web guy just said they might retry. So the damage cant be bad.
Other unoffical updates:
Sandstorm is out with a blown motor.
Sci II is supposedly still running.
Caltech is still running technically but is going nowhere fast. Its not stuck or anything tho.
Dad is running.
#25 has a stuck brake.
#23 has GPS problems but they may restart.
Navigator is stuck real bad in a fence. they are cutting it out.
#15 lost it hydralic pump
Cajunbot hit a obstacle right out of the gate.
Ensco hauled ass out of the gate, made its first turn OK but then rolled on the next.
Real driving involves seeing many things. Tablet PCs can't even read my handwriting if I write programming instructions.
There's also backtracking in case you can't find your way through a maze or roadblock.
Not to mention being able to ask for directions, finding fuel or requesting service.
How much brute force speed in terms of TFLOPS would be required?
Know your pads. One time pad: good for cryptography. Two timing pad: where to take your mistress.
Watching the live broadcast, the cyberrider just refuses to fight! The flagman waved the green go flag and instead of charging like a good infantryman it just threw itself down to the ground! Of course its from Berkeley so no that suprising I guess.
DARPA has announced that in light of the difficulties encountered this year, next's year event will be reworked as the "reasonably-ordinary-challenge' and consist of an autonomous vehicle locating the nearest McDonalds, ordering burgers and fries, and returning before they get cold.
What they really need to win in a race like this is a black mid 80s Pontiac Trans Am.
They also needed to widen that front sensor and put some HID LEDs in a chaser for effect.
Then hire William Daniels to replace all the chime codes with his voice.
Perhaps a pain in the butt to deal with, but a tracking client for Linux is available.
I work with autonomous robots in a lab setting. It's difficult enough to get a 2 wheeled robot the size of an RC car that moves 75 cm/sec to navigate its environment reliably. Failure is something you simply have to learn to live with and learn from (many computer scientists have a tough time getting used to the idea that these systems cannot nor ever will work 100% of the time). Honestly, you learn way more from failure than success in this business.
To get a full sized vehicle working at battlefield speeds with battlefield obstacles is a monumental challenge and almost certainly guaranteed to fail on the first try. Autonomous robotics is still a very young field, and the research published out there is generally some pretty rudimentary stuff done in a lab. Translating that stuff into a big complicated machine in a big complicated environment is a hell of a task and probably demonstrates some substantial holes in the current tech that weren't apparent from the confines of the lab.
This DARPA challenge does two excellent things for the field: Gives it a real goal and gives it a real deadline. Alot of research doesn't have a deadline and so researchers spend much of their time spinning their wheels (heh) on some of what i would consider, less important issues. This challenge gives a genuine goal to accomplish in a certain amount of time.
I definiately want to see the post-mortem on each team to see where they failed. In 2 more years, with this failure experience gained, perhaps a quarter of the teams will succeed or at least get further down the course.
-