Published on October 25th, 2016 | by Matt Pressman0
Inside NVIDIA’s New Self-Driving Supercomputer Powering Tesla’s Autopilot
October 25th, 2016 by Matt Pressman
Originally published on EVANNEX.
When Tesla Motors CEO Elon Musk announced new self-driving hardware, he said, “it’s basically a supercomputer in a car” capable of Level 5 autonomy after a sequence of software updates rapidly improves its performance over time. So who makes this supercomputer? NVIDIA announced that Tesla’s, “in-vehicle supercomputer is powered by the NVIDIA DRIVE PX 2 AI computing platform… [and it] delivers more than 40 times the processing power of the previous system. It runs a Tesla-developed neural net for vision, sonar, and radar processing.”
Love my new rocket! Congrats, Elon and the @TeslaMotors and NVIDIA teams. – Jen-Hsun Huang pic.twitter.com/KXdkuSMr3j
— NVIDIA (@nvidia) October 27, 2015
NVIDIA CEO Jen-Hsun Huang shows off his own (very admirable) Tesla car collection (Twitter: NVIDIA)
The CEOs of both NVIDIA and Tesla have a strong history of collaborating with one another. NVIDIA’s Huang is, himself, a Tesla owner and has been involved with the company as a key supplier with its Tegra chips. And, last year Huang invited and interviewed Tesla CEO Elon Musk at NVIDIA’s GPU Technology Conference. Most recently, as seen below, Huang donated “the world’s fastest AI system” to Elon Musk’s non-profit, OpenAI.
Obviously, the almighty DGX-1 (above) isn’t being used in Tesla’s cars. That said, what’s this new DRIVE PX 2 supercomputer for self-driving Teslas all about? NVIDIA explains, “DRIVE PX 2 is an end-to-end AI computing system that uses groundbreaking approaches in deep learning to perceive and understand the car’s surroundings… [and] lets carmakers first train their own deep neural networks on GPU supercomputers. Once loaded into the car, it processes the networks at high speed to provide the real-time, accurate response required for autonomous driving.”
Above: NVIDIA CEO Jen-Hsun Huang shows off the company’s new DRIVE PX 2 autonomous car technology at their CES 2016 press conference (Youtube: The Verge)
Here’s how Huang describes it (see video above) — think about the computational ability of 150 Macbook pros that all fits neatly into a kid’s school lunchbox. Furthermore, NVIDIA explains: “DRIVE PX 2 can understand in real-time what’s happening around the vehicle, precisely locate itself on an HD map, and plan a safe path forward. It’s the world’s most advanced self-driving car platform—combining deep learning, sensor fusion, and surround vision to change the driving experience… Use of Deep Neural Networks (DNN) for the detection and classification of objects dramatically increases the accuracy of the resulting fused sensor data.”
|NVIDIA Drive PX2 (Source: NVIDIA)|
We’ll be sure keep you posted on further news. But, it seems Wall Street is impressed thus far — Barron’s reports: “Shares of graphics chip titan Nvidia (NVDA) are up… after Tesla Motors (TSLA) apparently confirmed it is using Nvidia’s parts for its next wave of automation in its ‘Model S’ sedan and other models, throwing out technology from Mobileye (MBLY).” Indeed… goodbye MobileEye, hello NVIDIA.
Reprinted with permission.
Featured Image: Model X Kyle Field | CleanTechncia
Follow CleanTechnica on Google News.
It will make you happy & help you live in peace for the rest of your life.