Marco Polo Conference Room)
Virtually every automaker is working on driver assistance systems and self-driving cars. Conventional computer vision used for Advanced Driver Assistance Systems (ADAS) is reaching its limits as it is impossible to write code for every possible scenario in which a vehicle might navigate. Development of a truly autonomous car requires deep learning and artificial intelligence.
With deep learning, the vehicle can be trained to have super-human levels of perception in order to navigate more safely than humanly possible. Based on the same deep learning techniques used for detecting cancer and for beating the world champion Go player, artificial intelligence platforms are destined to enable fully autonomous vehicles.
This talk will cover the details of artificial intelligence in the automotive environment, and how a deep learning “supercomputer in a box” within the car itself will enable the dream of an autonomous vehicle. Also examined will be various implementations currently under development within the industry.