DARPA’s GRAND CHALLENGE DEMONSTRATES IMMATURITY OF ROBOT TECHNOLOGY
13 Mar 04. The dramatic finish to the much publicized £1m DARPA Grand Robot Challenge ended in disaster on Saturday as not one robot vehicle came close to completing the 142 mile course in the Mojave Desert.
The Chicago Register reported that seven miles was the best any team could muster with both the Red Team out of Carnegie Mellon University and SciAutonics II out of Thousand Oaks, California reaching that distance. Most of the 15 teams in the field failed to go more than a few hundred yards. All in all, it was a pretty humbling display for unmanned robot vehicle technology.
Some teams ran into obstacles. Others had equipment such as brakes or navigation systems break down. And two qualifying teams declined to even compete. In total, only seven vehicles managed to travel at least one mile.
DARPA had been looking for a team to traverse the 142 mile course in under ten hours to pick up the $1 million prize.
It’s hard to say that the millions upon millions of dollars put into the Grand Challenge went to waste. The government will have first pick of the technology it considers promising, and the event brought a lot of attention to the robotics field. Still, the reality of the event did not come close to meeting the hype surrounding it.
As far back as October of last year, several teams boasted that their vehicles had passed early tests. The Red Team even posted videos of its dazzling custom Hummer besting desert terrain.
In addition, DARPA did everything it could to help out the contestants. The 142 mile course was much shorter than original proposals stretching to 250 miles. The vehicles were also sent off in a time-trial format instead of all taking off at the same time as first planned.
In the end, however, the results were at the low end of expectations. Few thought any vehicle would actually complete the race, but many observers predicted that the Red Team and others would come close.
DARPA plans to hold the event again in 18 to 24 months, dangling the $1 million prize to winners one more time. In the meantime, a second $1 million challenge independent of government watch has been set up for September.
The failure of the Grand Challenge vindicates DARPA’s decision to delay the introduction of the ground vehicle robot element of the FCS program. The Grand Challenge was held specifically to trial existing robot technology available form industry and the outcome has proved that the technology is still in its infancy and needs more maturity to meet any chance of introduction in the FCS timescale.