Race promises more than one winner

JennyBond · Race promises more than one winner

It is a multi-million-dollar motor race across rough desert terrain, but the real prize could be fewer US casualties in Iraq and Afghanistan.

Gathered in the car park of a casino in Nevada's Mojave desert, 23 self-navigating robot vehicles are competing in the 2005 Grand Challenge - the world's only driver-less car race, organised by the Defense Advanced Research Projects Agency or DARPA, part of the US Department of Defense.

The stakes in the race are not small. The team which completes the gruelling 131.6 mile cross-country course in the fastest time will take home a $2m (£1.14m) cash prize.

The military need for vehicles which can drive themselves is most pressing in Iraq where US logistical and military convoys come under frequent attack by insurgents.

Drivers lucky enough to survive blasts from Improvised Explosive Devices still face the risk of being taken hostage by insurgents

As Greg Young of Team ENSCO said: "If we can take a logistics vehicle and drive it across an environment like that in Iraq and save lives that's worth a lot of technology investment."

In fact, Congress has mandated that one third of the US military's ground vehicles must be able to operate autonomously by 2015.

If that sounds unrealistic, consider the important military role played by unmanned arial vehicles (UAVs).

Although remote-controlled from bases all over the globe, UAVs like The Predator possess a high degree of autonomy necessary because of the time-lag in relaying commands.

Boulders, cliffs

But flying in the open skies is much easier than navigating desert littered with boulders and cliffs and man-made obstacles including tunnels and tank-traps.

The principal challenge competitors face is teaching the robot-cars to spot obstructions and calculate a route round them fast enough to let the cars travel above the minimum competitive speed of 15mph.

Another technical conundrum is how to ensure the vehicles do not "imagine" obstacles where none exist.

It adds up to a formidable challenge.

Last year all the competitors failed to complete the course, while the best distance was a mere 7.4 miles.

This year there is a lot more sponsorship and a lot more money behind teams, and most competitors believe at least one will complete the course.

But away from well-funded teams based out of universities or large engineering firms, the finalists are still a splendidly eclectic bunch.

Source