- Car can be put into 'autodrive' on familiar routes by tapping on an iPad
- Surroundings recognised using cameras and lasers built into car's body
- Car slows to a controlled stop if there is an obstruction in the road
- Driver can take back control by tapping on the brakes
By Liz Hull
|
If you thought cars fitted with sat-nav were high tech, then think again. Scientists at Oxford University have created a robot car which drives itself.
They say the technology could be installed into mainstream cars within 15 years and would dramatically improve safety and cut congestion on the nation's roads.
It means that futuristic vehicles could take over the daily commute or school run.
Scroll down for video
Car in control: An Oxford University scientists demonstrates their newly developed navigation system
Accept? Tapping on the brakes returns control of the vehicle to the driver
Named RobotCar, the vehicle is a specially adapted Nissan Leaf electric car, which has small cameras and lasers built into its chassis.
When the car is driven manually, the lasers and cameras act as its 'eyes' mapping a 3D model of the vehicle's surroundings, which in turn are fed into its computer 'brain' stored in the car's boot.
The car is then able to 'remember' roads and suburbs enabling it to drive itself along a familiar route. RobotCar prompts the driver, via an iPad on the dashboard, to see whether they want it to engage the autopilot and at the touch of the screen the car takes over the controls.
Another laser, mounted underneath the car's front bumper, constantly scans the direction of travel - around 13 times per second - for obstacles, such as pedestrians, cyclists or other cars up to 164ft ahead and in an 85 degree field of view.
If an obstacle is detected the car slows and comes to a controlled stop. Similarly, the driver can tap the brake pedal, like in current cruise control technology, to regain control from the computer at any time.
VIDEO See how sensitive the controls are on the iPad controlled car
The scientists behind the system say it is far superior to conventional satellite navigation because it is much more precise - within a fraction of an inch compared to a few yards with sat-nav - and the car does not need a satellite reference to know where it is.
At 5,000, the prototype is also considerably cheaper than any other currently in development, although the aim is to produce one for as little as 100 in the not too distant future.
Professor Paul Newman, of Oxford University's engineering science department, who is leading the project, said the technology would be especially valuable for drivers who perform regular routes, such the daily commute or school run.
'We are working on a low-cost auto drive navigation system that doesn't depend on GPS, done with discreet sensors that are getting cheaper all the time,' Prof Newman said.
'It's easy to imagine that this kind of technology could be in a car you could buy.
'Instead of imagining some cars driving themselves all of the time we should imagine a time when all cars can drive themselves some of the time. The sort of very low-cost, low-footprint autonomy we are developing is what's needed for everyday use.'
Already there are cars that can park themselves and last year Google announced its self-driving car saying that it had covered more than 140,000 miles on American roads.
Look, no hands! The car would take over the driving on familiar routes
The car can sense people in the road and stop when in autodrive mode
The Oxford car differs from Google's by having fewer sensors and relying more heavily on an on-board three dimensional map of streets.
The basic map could potentially be maintained by local councils or highway authorities and regularly updated by vehicles.
For insurance, legal and health and safety reasons, the RobotCar has so far only been allowed to drive on private roads around the scientists' Begbroke Science Park on autopilot. But they are currently in talks with the Department of Transport to see how and when it could be used on public roads.
The next stage of their research, which is part funded by the Engineering and Physical Sciences Research Council and is a collaboration between the university and sponsors Nissan, Guidance Navigation and Mira, will be to make the car recognise and understand complex traffic flows and make decisions on which routes to take.
The Department for Transport estimates that cost of congestion will rise to 24billion a year by 2025 so vehicles, like this prototype, could help alleviate some of those costs by avoiding jams and giving the driver time off to do other tasks.
Prof Newman added: 'While there's lots more work to do, 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.'
For more information visit www.robotcar.org.uk
No comments:
Post a Comment