Following the release of self-driving car disengagement statistics earlier this month, Apple today has released a brief whitepaper about its self-driving car efforts. The company specifically focuses on its “approach to automated driving system safety.”
As you might expect from a company as secretive as Apple, the paper is inherently vague. The company glosses over how it uses artificial intelligence and machine learning to make its “products and services smarter, more intuitive, and more personalized.”
Apple says that it is excited about the use of artificial intelligence and machine learning “in many areas, including transpiration.” It outlines the three benefits automated driving systems have:
Interestingly, Apple offers a breakdown of the standards to which it holds its self-driving car operators:
In particular, we believe that automated driving systems (ADS) have the promise to greatly enhance the human experience in three key areas: improving road safety, increasing mobility, and realizing broader societal benefits.
Apple also offers a very generic explanation of how its self-driving system works, saying it uses “a combination of sensors, including LiDAR, radar, and cameras.” This, of course, offers very little insight into how Apple’s system is different from systems of competitors.
Data from the California DMV earlier this month highlighted disengagement statistics for all companies testing self-driving cars in the state. This breaks down the number of times a human driver has to take over control of an autonomous driving system. According to the filing, Apple logged 871.65 disengagements per 1,000, for an average of 1.1 miles per disengagement.
If you’re interested, you can read Apple’s full 7-page filing with the National Highway Traffic Safety Administration here.
