NVIDIA Boosts IQ of Self-Driving Cars with World’s First In-Car Artificial

The new platform has been described by NVIDIA as the first in-vehicle artificial intelligence (AI) supercomputer in the world.

Nvidia has just announced a revamped version of its computing platform for self-driving cars, the Drive PX 2, which is claimed to deliver an enormous processing power equivalent to 150 MacBook Pros. Powered by 12 CPU cores with a combined throughput of eight teraflops and 24 Trillion deep learning operations per second, the Drive PX 2 is touted by Nvidia as the “supercomputer in a lunchbox”.

NVIDIA DRIVE PX 2 learns quickly to address the challenges of everyday driving such as construction zones, erratic drivers, and unexpected road debris.

Called the Nvidia Drive PX 2, the supercomputer is the size of a lunchbox and has processing power equivalent to that of “100 MacBook Pros”.

Revealing the system in Las Vegas today, Nvidia CEO Jen-Hsun Huang said simply that “self-driving cars are hard” – and his company knows how to get the job done.

As part of the self-driving vehicle equation, this computer needs to process imagery from multiple cameras, radar and LIDAR, laser-based sensor, transceivers, building a virtual world in real-time matching the computer’s external environment. And according to NVIDIA, the Drive PX 2 offers over 10 times more computational horsepower than the previous-generation product.

Nvidia also announced that Volvo would be the first vehicle company to deploy “several hundred cars” with Drive PX 2 to “start developing their self-driving capabilities”.

NVIDIA ultimate goal is to create an end to end platform for deep learning and autonomous driving, that incorporates a deep learning platform for training, deep neural networks, and in-car AI.

He added that the company’s GPU is key to the advances in deep learning and supercomputing. The unit features water-cooling, to enable operation in severe conditions. The ability to rapidly train deep neural networks on vast amounts of data is critical. Nvidia calls this mesh of self-improving “brains” the “Drivenet”.

The new Drive PX 2 is getting a serious hardware boost with the inclusion of two dedicated graphics processors (the Drive PX was based around two Tegra X1 SoCs), and that should allow automakers to really push what’s possible in real time and push the self-driving vehicle a bit closer to reality and final (self) drive-able products. According to NVIDIA, DriveWorks “enables sensor calibration, acquisition of surround data, synchronization, recording and then processing streams of sensor data through a complex pipeline of algorithms” that run on the Drive PX 2.