id="info"

CES 2016: Nvidia unveils new self-driving car tech with power of 150 MacBook Pros

In what has become an annual tradition of sorts at CES, Nvidia got the ball rolling ahead of the world's largest tech trade show this week with topics that will likely dominate the tech industry the year to come.

LAS VEGAS---In somewhat of a tradition at the annual Consumer Electronics Show, Nvidia exited the gates ahead of the official launch and opened by hinting at the major hot topics of the week.

Past examples have ranged from mobile gaming to the Internet of Things, but Nvidia's top brass touted its news this year as revolutionary on a global scale.

"We're going to talk about cars," quipped Nvidia CEO Jen-Hsun Huang immediately as he jumped out on the keynote stage at the Four Seasons Hotel on Monday evening.

Self-driving is a major computing science challenge, Huang admitted ahead of outlining an update on Nvidia's self-driving car strategy several years in the making.

Huang didn't waste time in unveiling the Nvidia Drive PX 2, a liquid-cooled super-computer with 12 CPU cores and four Nvidia-branded Pascal GPU chips to enabled self-driving cars.

Altogether, the hardware offers the power of 150 MacBook Pro laptops, Huang boasted, describing the entire super-computer fits in the trunk in the size of a standard school lunchbox.

The contributions to society, Huang opined, are revolutionary for the potential of self-driving cars to new mobility services and even redesigns for urban planning.

"Humans are the least reliable of the car," Huang argued, continuing that "we represent almost all of the fatalities around the car worldwide."

Huang acknowledged that "self-driving" technology is hard to achieve, but that didn't dissuade him from countering that "highway driving is relatively easy." It's city driving, he followed up, that present the most roadblocks to the innovative mode of travel.

Nvidia is tying its self-driving car tech -- arguably a flashier development for car makers and drivers alike -- back to another initiative at the top of the company's agenda: deep learning.

Earlier on Monday, Nvidia got the ball rolling early by unveiling plans for a program designed to bring PCs up to spec, thus equipping systems to better handle virtual reality.

The graphics card maker has been actively ramping up its machine learning and artificial intelligence portfolio over the last year.

Last spring, Nvidia unveiled several new technologies for advancing deep learning amid the GPU Technology Conference. Following up the Titan X platform for mobile gaming, the Pascal GPU series arrived with the promise to speed up deep learning applications tenfold compared to Nvidia's previous Maxwell processors.

In August, Nvidia bolstered its Grid platform for virtual desktops and applications with the debut of version 2.0, promising both the delivery of the most graphics-intensive apps ever as well as double both the performance and user density than its predecessor, now allowing up 128 users per server.

The tech titan then followed up in November with a new end-to-end hyperscale data center platform built especially for deep learning as well as Jetson TX1, a credit card-sized module promising to power millions of new smart devices to come.

Some of the machines Nvidia has in mind for Jetson TX1 include smart surveillance drones monitoring crowds for suspicious activity and robots that act like personal chauffeurs, recognizing its owners habits and needs through natural language, navigation and behavior recognition.

Tying all these threads together, Huang explained how these developments will connect an ecosystem rooted in an end-to-end platform for deep learning that will link the in-car AI super-computer to deep neural networks to the car itself.

The way Nvidia sees this loop, Huang explained, is that this "new world will be a cloud-connected Internet of Cars, Internet of Things, architecture" in which cars exit the line with "superhuman" features.

Over the last several months, Huang revealed Nvidia has been testing its own end-to-end system, now dubbed Nvidia DriveNet, for multiple and single object detection on the road. This is one of the most important parts of deep learning training, Huang stressed, because the entire platform is reiterating overtime.

"This network, because we want to put it into a car, has to recognize things in real-time," Huang noted, adding later that Nvidia is working to train its system to recognize both objects and circumstances, from undistinguished road lanes in snowy territories or emergency vehicles.

With a powerful enough computer and deep learning algorithms working on one software stack, Huang theorized, it's possible to imagine navigating yourself down the road in a self-driving car, safely.

Newsletters

You have been successfully signed up. To sign up for more newsletters or to manage your account, visit the Newsletter Subscription Center.
Subscription failed.
See All
See All