Apple driverless car research revealed

Apple computer scientists’ research on how driverless cars can better “see” cyclists and pedestrians while using fewer sensors has been posted online.

It could be the company’s first publicly disclosed paper on autonomous vehicles, Stuff reports.

The November 17 paper by Yin Zhou and Oncel Tuzel, used by online journal arXiv, shows the scientists propose a new software approach,”VoxelNet”, to assist computers in detecting three-dimensional objects.

Apple declined to comment, Stuff says.

The Apple Machine Learning Journal was established in July for its researchers and their work rarely appears outside it. Any of its research on self-driving cars has previously not been published.

Autonomous cars often use a combination of normal two-dimensional cameras and depth-sensing Lidar units to recognise the world around them. While the units supply depth information, their low resolution makes it hard to detect small, distant objects without help from a normal camera linked to it in real time.

But with new software, the Apple researchers say they were able to get “highly encouraging results” in spotting pedestrians and cyclists with just Lidar data. The experiments were computer simulations not involving road tests.

Apple filed a driverless car testing plan with California regulators in April.