Oxford University has unveiled a new autonomous car system that takes a different approach to Google's self-driving vehicles. RobotCar UK uses front-mounted 3D laser scanners and cameras on the vehicle to compare the environment to stored map data; if the car feels confident in its ability to take the wheel, it will notify the driver via software running on a dashboard-mounted iPad. Currently the system runs on a modified Nissan Leaf and has been tested at up to 40mph.
"It's not total autonomy for the car," says Professor Paul Newman, speaking to the Guardian. "It knows when things are good, and when the risks are reasonable, and then it will offer to take over." Unlike Google's system, RobotCar UK doesn't use GPS, and is restricted to environments it already knows. Still, it could work well for commuters that drive the same route every day; "Our cities don't change very much," says Newman, "so robotic vehicles will see familiar structures and say 'I know this route - want me to drive?'"
"There's no obvious legal barrier to using it on roads now."
RobotCar UK may offer certain advantages over Google's approach, too. The current array of technology costs around £5,000 (about $7,760), but the team hopes to get that down to £100 ($155) in the future. And, since the system is technically less ambitious, it could face less obstacles on its path to regulatory approval. "There's no obvious legal barrier to using it on roads now," says Newman. "It's essentially an advanced driver assistance system."