Apple researchers offer best look yet at self-driving program

It's believed to be the first public paper from Apple's secretive efforts.

Andrew Krok Reviews Editor / Cars
Cars are Andrew's jam, as is strawberry. After spending years as a regular ol' car fanatic, he started working his way through the echelons of the automotive industry, starting out as social-media director of a small European-focused garage outside of Chicago. From there, he moved to the editorial side, penning several written features in Total 911 Magazine before becoming a full-time auto writer, first for a local Chicago outlet and then for CNET Cars.
Andrew Krok
2 min read
Greg Baker/AFP/Getty Images

Apple's self-driving car program has been known for some time, but its research has remained somewhat mysterious -- until now.

Some of Apple's computer scientists have published a paper on 3D object detection, Reuters reports. The paper, which was submitted to the online journal arXiv, is believed to be the first time Apple has a publicly disclosed paper regarding its autonomous-vehicle development.

The paper, which you can check out at this link, discusses a new software program called VoxelNet, which helps detect three-dimensional objects with fewer sensors than many companies are currently using. VoxelNet managed to produce "highly encouraging results" through the use of light-based lidar alone, as opposed to linking lidar with another system, like a standard camera. The results come from the use of simulators, not Apple's on-road efforts.

Enlarge Image

Getting to peek behind Apple's curtain is a rare opportunity, especially when it comes to Project Titan.

Josh Edelson/AFP/Getty Images

Unless you have a degree in computer science, your eyes probably glazed over by the second page. What's important here, beyond simplifying sensor layouts used for object detection, is that this offers a rare peek into Apple's quite-secretive Project Titan.

Project Titan was originally supposed to be a whole self-driving car, but the tech titan eventually scaled back its efforts to focus on a platform alone, which it can develop and sell to automakers that lack the time, money or manpower to build a bespoke autonomous solution. This is the same route that Waymo, Uber and others are taking. Cars aren't cheap -- it would cost more than $1 billion just to engineer and certify the car itself, nevertheless its advanced self-driving systems.

While the program is no longer a secret, finding Apple testing in public has not been easy. Few sightings have been reported on -- a RX sporting Apple's system was spotted in May, and a second, more advanced sensor array was posted to Twitter in October.