3 Bots Win Pentagon's Robotic Rally

An anonymous reader writes "We've got a winner in the Pentagon's $3.5 million all-robot street rally, the Urban Challenge. Three, actually. Wired reports that 'bots from Stanford, Virginia Tech, and Carnegie Mellon all completed the course within the six-hour time limit. The robo-cars had to complete different missions taking varying times, so the flesh-and-blood judges will take a day to figure out who takes home first prize."

Actually, 4 potential

[ZaMoose]

Ben Franklin Racing [benfrankli...ngteam.org] (a collaboration between UPenn, Lehigh and Lockheed Martin) also finished within the 6 hour time limit.

The judging will certainly be interesting.

Re:Actually, 4 potential

[vertinox]

Yeah, but I think Ben was penalized for its initial delay at an intersection for too long at the first mission. After they rebooted it seemed to be ok and actually did pretty good getting around the intersections it had problems with. When Honeywell was taken out of the race, Ben was waiting behind it at the stop sign until they pulled Honeywell out but I don't think they penalize for that.

It is something to note that the two teams that finished but finished last were MIT and Cornell which had a collision with each other somewhere around Mission 2. But they both finished which is pretty awesome considering what it takes to run this corse.

Better Coverage

[Anonymous Coward]

That article is pretty sparse on detail. The best coverage I found was at

http://www.theregister.co.uk/2007/11/03/darpa_uc_blog/ [theregister.co.uk]

Re:We WILL have androids in 20 years

[LS]

I hear what you are saying - I also believe that anything is possible given enough time and hard work. Yet I think you are VASTLY underestimating the task of creating a human-like intelligence. Faster and more powerful != More Intelligent. Flight and chess are child's play compared to the human mind. It's also a false assumption to believe that a Turing architecture machine will be able to simulate the human brain with whatever specious equivalence used to compare human and computer processing power. The brain is NOT a computer. Computers themselves are simple expressions of a mere slice of how we understand our own mental processes to work. Do you know anyone who understands how any mind works, let alone their own, whether they be computer scientists, psychologists, cognitive scientists, or neuro-biologists? To put it simply, in order to expect a human-like intelligence in 20 years requires two things we do not yet have: An understanding of human intelligence, and a hardware architecture that is able to implement it.

LS

another reply

[SmallFurryCreature]

You mention chess. Alright, Deep Blue. Lets challenge deep blue. Half way through the game, we switch the board, introduce a new rule. Jumble the pieces up and tell it to pick them up and put them in the correct place again.

NOT a challenge for a human being. Deep Blue? Will fail totally, unable to even understand the commands.

AI is NOT the same thing as doing a simple task over and over again really fast. Laser range finders are nothing new, ACTING upon that information, THAT is the challenge. Especially when that information is not constant and reliable.

Kasparov showed that when he switched styles constantly, Deep Blue was unable to cope. That Kasparov went on to beat Deep Blue is often forgotten. It however showed very clearly that Deep Blue had been setup by HUMANS to beat Kasparov, when he became another player by changing style, the computer could not cope, it had no AI to deal with this.

It reminds me of futurama and robot blurns ball. Putting a howitzer on the field does NOT prove robots are good pitchers. IF Deep Blue could be put in front of a `checkers board and pick that game up in seconds, like a human could, then switch to TicTacToe and then play some poker ALL without human input, then I would be impressed.

Re:We WILL have androids in 20 years

[inca34]

This has little to do with Moore's and a lot to do with the fact that sensors do not follow Moore's law. We were using the same sensor technology as was available 15 years ago with marginal to no improvement on quality or capability.

The software side of this DARPA Urban Challenge should consist of no more than an enormous, but straightforward, state machine that contains all the logic for traffic decisions. Plug that into a simulator and you've got the main software part done.

The problem everyone had out in the field for the qualifiers (I was on one of the teams) was perception. How do you know what you see is an obstacle? And how do you deal with false positives, and more importantly, false negatives? Some people believe in cross referencing sensor data, which is called sensor fusion. It is difficult, to say the least, to characterize every possible obstacle that ought to be considered a true obstacle if it lies on your vehicles path, let alone have a 10^-6 failure rate for improper detection.

The highway lane following has been solved since the 70s, check out R.E. Fentons work on Automated Highways in Transportation Science (1970). We had some "recent" developments in the early 1990s where we got some autonomous vehicles to do the autobahn at 100mph with more modern sensors and vehicles, but really didn't improve that much because the sensors aren't there yet.

Your sensor choice goes something like this:
$75k for a Velodyne 3D laser system
$5k for the SICK 2D (planar) lasers
~$25k for stereo vision cameras (per set)
~$1k for radar
$75K for the Applanix integrated GPS and IMU

The Velodyne is a spinning set of 64 lasers, with 64 photodiodes. Each manually placed so that the photodiodes are aimed precisely where the lasers are pointed. The entire head of the unit spins at ~2Hz and generates 1 million points per second. Most of the teams that bought one mounted it on top of their vehicle. This sensor is great if you have infinite processing power available to crunch the data and turn it into cost maps. It however has some serious problems: it's very expensive, it's not mass manufacturable, the point data for a rock and a shrub are indistinguishable (a weakness of all lasers), some obstacles we're interested in absorb laser or reflect it away from the photodiodes, it has too much information, and it has moving parts.

The SICK 2D planar lasers have more or less the same problems, except there's less data to crunch of course. These lasers also have moving parts internally, which spin a mirror at maybe ~20Hz to get distance data over a 2D plane. Same issues as the Velodyne, except it's manufacturable (has been for 15 years now).

Stereo vision is really hard to do right. When you have roughly a year to develop the platform and the algorithms, I don't expect much, and I didn't see much. This may be the answer in the future for passive detection, but I don't see it working at the moment.

RADAR is the right sensor for this type of work. It gives you distance and speed. If you're clever it also gives you the "cost" of a particular object. Radar is how you can tell the difference between a shrub and rock, or a car and a plastic fence. The real cost in the RADAR is not the sensor, but the $100k guy who knows RADAR well enough to set it up right and get good data out of it.

The Applanix GPS and IMU with 200k RPM laser gyros are not manufacturable and not practical for autonomous vehicles because of the cost. Perhaps the MEMs solutions will catch up and make IMUs cheaper, but in the mean time we're stuck with these systems if you care about your position.

That's my take on it. Improve the sensors and we'll get autonomous vehicles. Buy another Cray, strap on a generator and a multi-ton air conditioner is not the solution. We need to reliably and cheaply generate cost maps that are relevant to the vehicle that's being automated. Once that's been done reliably, we will have autonomous vehicles. Cheers.

Re:We WILL have androids in 20 years

[ucblockhead]

Lots of people predicted cell phones. In Heinlein's futurist essay written in 1950, he predicted in 2000 that everyone would have a wireless phone you could put in a pocket. He revisited this essay a couple times and in the last revisitation in 1980 he referred to the wireless phone prediction as "obviously correct".

Re:Actually, 4 potential

[ZaMoose]

BFR's blog actually makes [blogspot.com] that claim (sub-6 hours).

The winners have been announced

[SnowZero]

On the official website [darpa.mil].

1. Tartan Racing (Carnegie Mellon)
2. Stanford Racing Team (Stanford)
3. Victor Tango (Virginia Tech)