Researchers at the University of Oxford have developed a modified Nissan Leaf that can drive itself — based on its recognition of its surroundings.
The RobotCar U.K. project used the Nissan Leaf, an all-electric vehicle that will cost customers $28,800 after its latest price drop in January, fitting it with around $7,750 worth of equipment.
A team of 22 researchers started work on the project in September, working alongside Nissan to develop it. "We are not condemned to a future of congestion, accidents and time wasting," RobotCar U.K.'s website declares on its front page.
A PC controller sits in the trunk, with cameras in the front to watch for pedestrians and other obstacles, lasers under the front and rear bumpers and an iPad for the user interface, reports GigaOM. The developers hope to eventually build a system that costs only £100, which translates to about $155. Paul Newman, the co-leader of this project, said in a statement that the system would not depend on GPS because the satellites are not accurate enough for the researchers.
The car's functionality is based on the machine's ability to learn as it maps the city and its streets, recognizing landmarks and familiar structures as it goes. Twin cameras will watch the road while lasers create a three-dimensional map of the car's environment, which is similar to Google's system, although the search engine giant's LIDAR unit would cost $70,000. Once the system recognizes its surroundings, the car will inform the driver that it recognizes its surroundings and ask if it should take over. Rather than a fully autonomous car, this system can easily switch back to user control if the driver taps the brake pedal.
The idea of a self-driving vehicle has been around since 2010, when Google first began work on its autonomous car. A computerized car would be more fuel-efficient, optimizing speed and routes, and would potentially cause fewer accidents than someone sitting behind a wheel who occasionally gets careless. Fewer accidents means less need for heavy-duty safety features, and cars that are more lightweight.
How long before a system like this is eventually refined and adopted into the mainstream is still up in the air — after all, people still have trouble switching to hybrids and better fuel standards. It will be years before the system will be in a showroom, says Newman, but he adds, “It shows the potential for this kind of affordable robotic system that could make our car journeys safer, more efficient and more pleasant for drivers.”