Nvidia revealed a new artificial intelligence computer based on its Drive PX platform designed specifically to bring Level 5 (completely self-driving) autonomous vehicles to market. The Drive PX Pegasus builds on its predecessor the Drive PX 2, offering over 10 times as many operations per second in terms of computing capability, and it's already being used by over 25 companies to develop fully driverless taxi vehicles for eventual deployment.
The self-driving computer is built on Nvidia's CUDA GPUs, and also increase the practicality of the computing hardware requirements for actually fielding self-driving cars on real roads. Nvidia points out that vehicles today using Nvidia tech to field on demand autonomous taxis include basically entire data centres in the trunk, and Drive PX Pegasus is designed to help put them on track to bring this down to something more manageable for actual production deployment.
Drive PX Pegasus packs in four AI processors, including two of Nvidia's latest generation Xavier system-on-a-chip models with embedded Nvidia Volta GPUs, with two next-generation discrete GPUs designed specifically to hep accelerate deep learning and computer vision. The overall size of the system is roughly that of a license plate, Nvidia says, and it greatly reduces energy consumption and overall cost as well.
Pegasus is set to be available to Nvidia partners during the second half of 2018, while companies can get started now with Drive PX 2 if they want to chart a path to upgrade later on.