A driverless car system under development at Apple has been showcased to a select group of machine learning experts.
The firm’s director of AI is reported to have shared secret details of its ongoing automated motoring projects at an industry event.
This is believed to included the tech company’s self-driving technology that uses laser sensors, called ‘VoxelNet’, to spot cyclists and pedestrians.
A driverless car system under development at Apple has been showcased to a select group of machine learning experts. The firm’s director of AI is reported to have shared secret details of its ongoing automated driving projects at an industry event
The presentation was given by Ruslan Salakhutdinov during the Neural Information Processing Systems conference, being held in Long Beach, California.
Nearly 8,000 AI researchers have attended the event, where Apple gave a sneak peak into some of its ongoing projects.
Mr Salakdinov, who joined Apple in 2016, gave a demonstration of the VoxelNet technology, as well as a number of other yet to be announced projects.
This included a method for identifying different objects on the road using cameras on top of a self-driving vehicle.
‘If you asked me five years ago, I would be very skeptical of saying “Yes you could do that”,’ according to reports in Wired.
It also delivered information about a technique, known as visual simultaneous location and mapping (Slam), which uses camera footage to more accurately track a vehicle’s exact location.
Though chief executive Tim Cook has called self-driving cars ‘the mother of all AI projects,’ Apple has given few hints about the nature of its self-driving car ambitious.
‘VoxelNet’ was first revealed in a paper, submitted on November 17 to independent online journal arXiv, by Yin Zhou and Oncel Tuze.
It is significant because Apple’s famed corporate secrecy around future products has been seen as a drawback among artificial intelligence and machine learning researchers.
The scientists proposed a new software approach for helping computers detect three-dimensional objects.
Academics are used to freely sharing their work with peers at other organisations.
Self-driving cars often use a combination of normal two-dimensional cameras and depth-sensing ‘LiDAR’ units to recognize the world around them. But a new software approach for helping computers detect three-dimensional objects has been developed by Apple
Yielding to that dynamic, Apple in July established the Apple Machine Learning Journal for its researchers.
Their work rarely appears outside the journal, which so far has not published any research on self-driving cars.
Self-driving cars often use a combination of normal two-dimensional cameras and depth-sensing ‘LiDAR’ laser units to recognise the world around them.
While the units supply depth information, their low resolution makes it hard to detect small, faraway objects without help from a normal camera linked to it in real time.
But with new software, the Apple researchers said they were able to get ‘highly encouraging results’ in spotting pedestrians and cyclists with just LiDAR data.
They also wrote they were able to beat other approaches for detecting three-dimensional objects that use only LiDAR.
The experiments were computer simulations and did not involve road tests.
Last December, Apple told federal regulators it was excited about the technology and asked regulators not to restrict testing of the technology.
In April, Apple filed a self-driving car testing plan with California regulators.
Though Chief Executive Tim Cook has called self-driving cars ‘the mother of all AI projects,’ Apple has given few hints about the nature of its self-driving car ambitious