Apple’s ambitions in the self-driving car segment is pretty well known to us. A group of Apple scientists have now published a research paper on autonomous car tech that talks about judging obstacles on the path using just LiDAR (Light Detection and Ranging) sensors and no cameras.
Conventional self-driving cars come with a combination of LiDAR sensors and standard cameras. But the scientists working for Apple have devised a way to judge pedestrians and cyclists on the car’s path using just LiDAR sensors. The paper was published by Yin Zhou and Oncel Tuzel in a paper titled “End-to-End Learning for Point Cloud Based 3D Object Detection.” It is important to note that the tests were only conducted on computer simulations and haven’t been tested on the roads yet. However, the scientists seem optimistic about its results, and this could well prove to be groundbreaking.
Apple has eased its privacy policies over the past year or so and is sharing more details on its research. Apple CEO Tim Cook famously mentioned that autonomous tech is “the mother of all AI projects” which is an indication of things to come. The company has hinted that autonomy can be used in other segments and won’t just be limited to cars or other vehicles.
While Apple was initially believed to be working on a self-driving vehicle, the company’s ambitions have now shifted onto self-driving tech. About a year ago, Apple mentioned that it will allow researchers to share machine learning papers to the public. We can expect more research to be shared as they become available.