Nvidia Unveils Single-Chip Version of Drive PX 2 for Self-Driving Cars

The new offering will expand the GPU maker's portfolio of products aimed at the rapidly growing autonomous vehicle market.

Nvidia cars

Nvidia officials are looking to accelerate the development of self-driving cars with the release this week of a smaller version of the Drive PX 2 platform that was introduced earlier this year.

At the company's GPU Technology Conference (GTC) in Beijing Sept. 12, Nvidia executives introduced the new single-processor Drive PX 2 that is designed for what they called "auto cruise" functions, including automated driving and high-definition mapping. Like its larger brethren, the new version uses artificial intelligence (AI) capabilities and machine learning to enable the computer system in the car to collect, storage and process data from multiple cameras and sensors in real time.

Through the capabilities of Drive PX 2, self-driving cars will be able to instantly understand the environment around them and react accordingly to avoid accidents and enable point-to-point autonomous driving, according to Nvidia officials. The new version of the platform will make it easier for car makers to incorporate the Nvidia technology into their designs, according to Rob Csonger, vice president and general manager of Nvidia's automotive business unit.

"Bringing an AI computer into the car in a small, efficient form factor is the goal of many automakers," Csonger said in a statement. "Nvidia Drive PX 2in the car solves this challenge for our OEM and tier-one partners and complements our data center solution for mapping and training."

The palm-sized computer will be used by Chinese tech vendor Baidu as the in-vehicle car computer for the company's map-based self-driving offer to car makers. Nvidia and Baidu announced a partnership in August in which the company's will unite Baidu's cloud and deep-learning capabilities with Nvidia's self-driving car platform.

The new version is smaller than the Drive PX 2 system that was first in introduced earlier this year at the 2016 Consumer Electronics Show (CES) and later talked about at Nvidia's GTC show in April in California. That system uses Nvidia's Tegra "Parker" processor, which includes two of the vendor's "Denver 2.0" CPUs and two Pascal-based GPUs designed to power deep-learning applications that will make cars smart enough to recognize and respond to obstacles.

The newer version includes a single six-core CPU and a Pascal-based GPU. Nvidia officials have pointed to autonomous cars, AI and deep learning as among the emerging markets that the company is pursuing, seeing them as good fits for its GPUs. The new version of the Drive PX 2 gives automakers and suppliers a range of options—with configurations of one or two mobile processors to two discrete GPUs to multiple Drive PX 2s—that can support everything from auto-cruise and auto-chauffeur for point-to-point trips to a fully self-driving car, officials said.

Nvidia also offers its DriveWorks suite of software, which includes tools, libraries and modules, to help with the development and testing of autonomous vehicles.

The single-processor Drive PX 2 will be available to production partners in the fourth quarter. The dual-processor version and DriveWorks software are available now.

Nvidia is one of a number of chip makers looking to gain traction in the self-driving car space. Intel is making a significant push, and in July announced a program with BMW and Mobileye to develop a technology platform that will enable the automaker to put self-driving cars on the road by 2021. Meanwhile, NXP Semiconductors—helped by its acquisition last year of Freescale Semiconductor—is rolling out products for the space, and Renesas Electronics on Sept. 13 announced a deal to buy fellow chip maker Intersil for $3.2 billion in an effort to expand its presence in the automotive space.