We’re probably not going to see an iCar from Apple any time soon. Instead, the company is working hard on hardware and software to control the vehicles of other manufacturers autonomously. Apple is backing SLAM technology, Simultaneous Localization and Mapping. With this method, the software creates a map of the surrounding area and estimates its own position on this map.
The data required to do this come from the optical distance and speed measurement system Lidar. Within a three-dimensional map of the surroundings, the vehicle can recognise pedestrians, cyclists and traffic signs and adapt its driving style to the patterns of movement detected and even to those predicted. Even road users hidden by parked cars are to be recognised at the edge of the road. In a further sub-project, Lidar is supported by cameras. A new image recognition software is supposed to ignore interfering factors such as drops of water on the lens.
Insight into research process
Although it’s a secret as to exactly how the complex technologies work and interact, the company is being unusually open in for the first time granting a glimpse of its research process. In December 2017, Ruslan Salakhutdinov, Apple’s specialist for Artificial Intelligence, presented the progress made on the project at a congress. By doing so, the company doubtless wants to attract talented developers for whom it has to fight with competitors such as Facebook and Google. The more openly scientific experts swap notes on their work and are also allowed to publish, the more attractive a research project becomes for them.