Apple’s director of artificial intelligence research has outlined the software capabilities needed for self-driving cars, including using date from cameras and other sensors to spot cars and pedestrians on urban streets.
Wired Magazine reports that Ruslan Salakhutdinov told around 200 AI experts about how Apple uses machine learning to analyse large stockpiles of data and also how Apple is learning how to teach cars to navigate in unfamiliar spaces, and build detailed 3-D maps of cities.
“The talk offered new insight into Apple’s secretive efforts around autonomous-vehicle technology,” the report says. “The scale and scope of any car project at Apple remains unclear. Salakhutdinov didn’t say how the projects fit into any wider effort in automated driving, and a company spokesman declined to elaborate.”
It adds that Salakhutdinov showed data from one project previously disclosed in a research paper posted online last month to train software to identify pedestrians and cyclists using 3-D lidar scanners used on most autonomous vehicles.
The report says that other projects Salakhutdinov discussed don’t appear to have been previously disclosed. One created software that identifies cars, pedestrians, and the driveable parts of the road in images from a camera or multiple cameras mounted on a vehicle.