Nvidia has made a tiny artificial intelligence computer for powering the high definition mapping and highway automated driving of autonomous vehicles.
Snapped up by China's Baidu as the in-vehicle car computer for its self-driving cloud-to-car system, the palm-sized Drive PX 2 allows cars to use deep neural networks to process data from multiple cameras and sensors — all while using just 10 watts of power.
It promises to speed up production, too. A car using the Drive PX 2 for AutoCruise can understand in real time what is happening around it, locate itself on an HD map and plan a safe path forward.
The computer is powered by the company's newest "system-on-a-chip", featuring a GPU based on the Nvidia Pascal architecture. A single Nvidia Parker system-on-chip (SoC) configuration can process inputs from multiple cameras, plus lidar, radar and ultrasonic sensors. It supports automotive inputs/outputs, including ethernet, CAN and Flexray.
"Bringing an AI computer to the car in a small, efficient form factor is the goal of many automakers," said Rob Csongor, vice president and general manager of Automotive at Nvidia. "Nvidia Drive PX 2 in the car solves this challenge for our OEM and tier 1 partners, and complements our data centre solution for mapping and training."
Nvidia says more than 80 automakers, tier 1 suppliers, startups and research institutions developing autonomous vehicle solutions are using Drive PX. It comes in a mobile processor configuration, a combination of two mobile processors and two GPUs and multiple Drive PX 2s — so it can be used in everything from AutoCruise for the highway, to AutoChauffeur for point to point travel, to a fully autonomous vehicle.
It was announced last week that the Drive PX 2 will be the AI engine of the Baidu self-driving car.
"Baidu and Nvidia are leveraging our AI skills together to create a cloud-to-car system for self-driving," said Liu Jun, vice president of Baidu. "The new, small form-factor Drive PX 2 will be used in Baidu's HD map-based self-driving solution for car manufacturers."
Supporting other AI research Nvidia is conducting, data scientists who train their deep neural networks in the data centre on the Nvidia DGX-1 can run on Drive PX 2 inside the vehicle. The same Nvidia DriveWorks algorithms, libraries and tools that run in the data center also run in the car.
"This end-to-end approach leverages Nvidia's unified AI architecture," Nvidia says, "and enables cars to receive over-the-air updates to add new features and capabilities throughout the life of a vehicle".
The Drive PX 2 will be available to production partners in the fourth quarter of 2016, and the DriveWorks software and the Drive PX 2 configuration with two SoCs and two discrete GPUs are available today for developers working on autonomous vehicles.