Nvidia goes all-in on self-driving cars, including a robotic car racing league

Nvidia goes all-in on cocky-driving cars, including a robotic car racing league

This site may earn chapter commissions from the links on this page. Terms of use.

Jen-Hsun Huang raving about the Nvidia powered Robocar for Roboracing

Nvidia doesn't always announce new consumer graphics cards at its annual technology conference, simply it was widely expected to this year. Instead, GTC 2016 is all most AI, VR, and especially self-driving cars. Following up on its declaration of the Drive PX 2 car computer, Nvidia updated its plans to ship a complete set of developer tools — fueled past its own autonomous vehicle research — for car makers, and to sponsor and help equip a robot car racing league.

DriveWorks is the power behind the Drive PX 2

A supercomputer in your trunk, like Nvidia's Drive PX two, isn't much skilful without the software to run it. That'south where Nvidia's DriveWorks platform comes in. Kickoff announced at CES, it is getting closer to reality with a "Spring 2016" ship date. Nvidia CEO Jen-Hsun Huang also used his keynote to become into more detail about what information technology will include. The developer platform starts with sensor fusion and computer vision software that can work with up to 12 cameras and other sensors to provide a comprehensive model of the vehicle's environment. From there, advanced machine learning capability will assist with navigation, vehicle command, and path planning.

Path planning is the tricky process of deciding where to navigate the car in traffic or through an intersection Loftier-quality maps, similar those from Here, are likewise going to be supported. One interesting feature is back up for map creation using the DriveWorks in-car platform coupled with cloud-based processing for the actual map cosmos. It was a little unclear from Huang'southward clarification exactly how all this would work — except that he hopes and expects that that cloud will be populated with Nvidia'southward new $130K DGX-ane supercomputer — but what is clear is that he sees this applied science greatly reducing the cost of mapping areas, and of training autonomous vehicles. In particular, it should make it possible to do a better chore of keeping maps up to appointment. Instead of needing routes to exist re-driven with expensive, specialized, vehicles to pick up changes in the route layout or obstacles, information from "regular" Drive PX two-equipped cars could be used.

Self-driving with DAVENET, or "I can practise that, Dave"

Rounding out Nvidia's DriveWorks offering will exist a deep neural network (DNN) that has been trained to know how to drive. Traditionally, autonomous vehicles, such as the ones used in the DARPA challenge, have relied on manually-coded algorithms to follow a desired road, and provide vehicle control. Nvidia (forth with many other current vehicle research teams) has been experimenting with using deep learning neural networks instead. According to Huang (and illustrated with a demo video), later only 3,000 miles of supervised driving, its motorcar — powered past its DAVENET (formerly named DRIVENET) neural network — was able to navigate on freeways, land roads, gravel driveways, and in the pelting.

The computing power required to recoginize and accurately track dozens of objects across many cameras and sensors is definitely supercomputer worthy Of course, what he showed was only a demo video. But all in all, it was quite a remarkable achievement when contrasted with the hundreds of homo years of coding that went into the much-less-sophisticated driving of the DARPA claiming cars only 10 years ago. Obviously, Nvidia isn't suddenly planning to become a car company, simply it will be providing its technology as function of the set of tools for the automobile industry to use to take advantage of its Drive PX ii. Huang showed, for example, how the PX ii'due south ability to process 12 cameras at one time not only assists driving safely through traffic and obstacles, but builds a sufficient model of the world around it to let for adjusting to road atmospheric condition and routing.

Roborace: Full-size robotic car racing

For decades, car and auto accessory manufacturers have used racing every bit both an advertising tool and a way to advance their own research and evolution. Whether it is F1, IndyCar, or NASCAR, mill teams are e'er present and always using what they acquire to assistance them with their next generation of street vehicles. Now that autonomous operation is an increasingly realistic future path for road cars, bringing computing front and heart in auto evolution, it makes sense racing should become a platform for AI-based vehicle R&D.

Nvidia's Drive PX 2 replaced a supercomputer in Baidu's autonomous vehicle project

That'southward exactly what Nvidia and others are planning for the newly appear Roborace league. Piggybacking off the fast-growing Formula E (all Electrical) schedule and machine blueprint, the league volition feature twenty identical Roborace cars allocated to x teams. They will race on the aforementioned courses as Formula E, except without drivers. The cars won't be remote-controlled, either. They'll be fully democratic, using an Nvidia Bulldoze PX 2 portable supercomputer to run their software. And so the teams' innovation and differentiation will be in the software they develop for the race. The Roborace is scheduled to start aslope the 2016-2017 Formula E season, subsequently this year. Roborace founder Dennis Sverdlov told GTC attendees he expected information technology to make heroes out of software developers: "Information technology's not possible to get competitive reward based on how much money you put in hardware. Our heroes are non the drivers. Our heroes are engineers."

Jealous? You too can build a (minor) cocky-driving car!

You can DIY your own robo racecar by following along with Jetsonhacks Along with each new autonomous vehicle announcement, in that location is always a statement of the massive investment needed to go far happen. But for those of us who want to practise more than than be passive spectators, there is an exciting new opportunity to larn how to build your own — scaled-downwards — robotic race motorcar. Startup JetsonHacks has taken MIT's RACECAR autonomous car learning platform and made it attainable to the DIY community with detailed assembly instructions, and cost-saving hardware options to brand it more than affordable than the University's original version. The RACECAR is a massive kit bash of an offf-the-shelf RC vehicle — a Traxxas Rally — so that all the DIY fun is concentrated on the command and programming. The brain is (naturally) a Jetson TK1, running Robot OS (ROS).

In an exclusive interview, JetsonHacks Founder Bill Jenson excitedly explained that this year will characteristic an upgraded model based on this Jump'south MIT Controls Course — which volition be bachelor online — and a new design featuring a more-powerful Jetson TX1. If y'all'd rather flex your maker muscle with a drone, he also offers a lot of great DIY drone advice based on the DJI Matrice 100 development platform.