Robotica Exotica

The Smart People Behind the Grand Challenge

The Grand Challenge seems to have struck a nerve with the rest of the media, who are all writing articles about the people behind the vehicles. The New York Times has this nice peice on the Return of AI:

At its low point, some computer scientists and software engineers avoided the term artificial intelligence for fear of being viewed as wild-eyed dreamers.

But the work of a small team of researchers at the Stanford Artificial Intelligence Laboratory is helping to restore credibility to the field. The team's winning robotic Volkswagen, named Stanley, covered the unpaved course in just 6 hours and 53 minutes without human intervention and guided only by global positioning satellite waypoints.

The feat, which won a $2 million prize from the Pentagon Defense Advanced Research Project Agency, was compared by exuberant Darpa officials to the Wright brothers' accomplishment at Kitty Hawk, because it was clear that it was not a fluke. Twenty-two of the 23 vehicles that started this year did as well or better than the seven miles completed by the best vehicle last year.

The ability of the vehicles to complete a complex everyday task - driving - underscores how artificial intelligence may at last be moving beyond the research laboratory.

While artificial intelligence technology is already in use in telephone answering systems with speech recognition and in popular household gadgets like the iRobot vacuum cleaner, none of the existing systems have been as ambitious as Darpa's Grand Challenge road race.

This leap was possible, in large part, because researchers are moving from an approach that relied principally on logic and rule-based systems to more probability or statistics-oriented software technologies.

"In the past A.I. has been dominated by symbolic systems and now the world is gray," said Terrence J. Sejnowski, head of the computational neurobiology laboratory at the Salk Institute in La Jolla, Calif. "That's what it's like to deal with the real world."

As they say, read the whole thing.

Stanford Wins the Grand Challenge

CONGRATULATIONS!!! and $2 Million go Stanford for winning the DARPA GRAND CHALLENGE, with an overall speed on the course of 19 MPH.

Several articles are all over this race, and the speed of progress from last year to this year is amazing. Last year, the best showing was 7 miles. This year, 21 of 23 racers went farther than 10.

Not so long ago, the notion of cars driving themselves seemed no more than science fiction. This weekend, it took a giant stride towards becoming reality as a handful of robotically controlled vehicles completed a 132-mile obstacle course around the Mojave desert.

Lured by a $2m (£1.15m) prize offered by the Pentagon, teams from some of the United States' leading universities threw themselves into the challenge and watched their souped-up four-wheel drives and military vehicles negotiate tunnels, lake beds, a steep mountain pass and numerous rocks.

First across the finish line was a converted Volkswagen Touareg developed by engineers from Stanord University - the prime research institution in Silicon Valley which was previously the springboard for many of the innovations of the internet.

"The impossible has been achieved," Stanford's team leader, Sebastian Thrun, crowed as the vehicle, nicknamed Stanley, completed the course in a brisk seven and a half hours. "The dream of cars driving themselves is becoming a reality. Before, the question was whether it was possible. Now we know it is."

Posted by elkaim at 10:24 AM

And They're Off!

The second annual DARPA GRAND CHALLENGE has begun in Primm, Nevada. This years race of autonomous vehicles across the desert looks to be more competitive than last years. Smart money is on the Stanford/VW team with the modified Tuareg, which performed very well in the NQE. Don't discount Red Team from CMU, either, as they are the other front runner. Rummors have it that they both completed last years run during testing.

The race is organised by the Pentagon's defence agency to push research into autonomous vehicles for the US military. This year it has doubled the prize fund for the challenge.

The vehicles have been kitted out and modified with GPS (global positioning satellite), cameras, infrared sensors, computing equipment, and lasers to guide them across the tough terrain.

They are not allowed to be controlled, even remotely, by humans.

"We can now see a future where these vehicles will take the place of soldiers in harm's way," said Ron Kurjanowicz, manager of the Darpa Grand Challenge race.

The captain of the Stanford University team, Sebastian Thrun, was confident that this year's race would produce a winner.

He was also confident about the future for autonomous vehicles. "It's a no-brainer that 50 to 60 years from now, cars will drive themselves," he said.

Tough heats

The 23 finalists were chosen after eight days of qualifying events over much shorter courses.

They include a Hummer built by Carnegie Mellon University, called H1ghlander, a converted Humvee named Sandstorm, a modified Volkswagen Touareg by Stanford University, a six-wheel truck and a Jeep Grand Cherokee called Spirit.

The 40-pupil team from Palos Verdes High School in California are in the running with their Doom Buggy modified SUV.

They were one of 195 teams that originally applied to take part in the gruelling cross-country challenge.

The 23 will have 10 hours to complete the race, which will include a human-made obstacle course.

But the precise route is kept secret until two hours before the competition.

The teams have had varying levels of sponsorship to develop their desert robots.

Some have had millions pumped into the projects from corporate sponsorship, while others have scraped together much less funding.

If no-one wins this time round, Darpa said it would most likely run another race.

See also the DARPA GRAND CHALLENGE site for more info.