Nvidia's AI Platform Could Enable Fully Autonomous Vehicles

Nvidia's AI Platform Could Enable Fully Autonomous Vehicles


The NVIDIA DRIVE PX "Pegasus" AI computing platform is the world’s first designed to drive fully autonomous robotaxis.

Autonomous vehicles could reshape the fabric of society, and everyone from Volvo to Google has thrown their hat in the self-driving race (to varying degrees of success). But Nvidia’s new Pegasus AI computing platform, which crunches 320 trillion+ operations per second, could finally make Level 5 driverless vehicles a reality.

The Society of Automotive Engineers (SAE) defines a “Level 5” driverless vehicle as one that “can handle all driving tasks and go anywhere. No human, no steering wheel, no pedals. Climb in, tell it where you want to go….”

It’s the holy grail of transportation technology.

"Creating a fully self-driving car is one of society's most important endeavors," said Jensen Huang, NVIDIA founder and CEO. "Driverless cars will enable new ride- and car-sharing services. New types of cars will be invented, resembling offices, living rooms or hotel rooms on wheels … the future of society will be reshaped," he said.

“Millions of hours of lost time will be recaptured by drivers as they work, play, eat or sleep on their daily commutes,” notes Nvidia’s press release.

This is all theoretical, of course, though 225 companies are tinkering with the Nvidia Drive PX platform, and more than 25 are using the onboard tech and computing power to develop fully autonomous robotaxis.

And this can’t be overstated – the Nvidia Pegasus is fast. With combined memory bandwidth exceeding 1 terabyte per second, the system should be able to handle all the computational demands of a fully autonomous vehicle (within a restricted geographic area like a city).

Of course, no one should trade in their “dumb” cars just yet – we don’t have “Level 3” cars on the market yet, and while Pegasus will be available to Nvidia’s partners in the second half of 2018, turning raw horsepower into a driverless horse will take time.

Read more here: https://nvidianews.nvidia.com/news/nvidia-announces-world-s-first-ai-computer-to-make-robotaxis-a-reality