NVIDIA: Drive PX Pegasus AI Computer Powers Level 5 Autonomous Vehicles

NVIDIA introduced its Drive PX Pegasus AI computer that it claims can power Level 5 autonomous vehicles. Will Pegasus help show a path to production for the automakers and tech companies working on fully autonomous vehicles?

Photo Caption: NVIDIA introduced its Drive PX Pegasus AI computer that it claims can power Level 5 autonomous vehicles. Will Pegasus help show a path to production for the automakers and tech companies working on fully autonomous vehicles?

NVIDIA introduced at its GTC Europe event in Germany the NVIDIA Drive PX Pegasus AI computer, which the company claims can power Level 5 autonomous vehicles.

According to NVIDIA, Pegasus can handle more than 320 trillion operations per second, which is more than ten-times the performance of its predecessor, the Drive PX 2. NVIDIA founder and CEO Jensen Huang said this type of performance is equivalent to the AI performance of a 100-server data center.

Pegasus will be available to NVIDIA partners in the second half of 2018. NVIDIA said there are 25 companies using NVIDIA’s chips to develop fully autonomous robotaxis, including nuTonomy, Yandex.Taxi and Zoox. NVIDIA said it expects fully autonomous vehicles to initially be rolled out in ride-hailing scenarios.

NVIDIA appears to have done a tremendous job in reducing the size of this type of system. Self-driving test cars have historically had trunks full of computers, but Pegasus, NVIDIA said, is about the size of a license plate.

These types of computers, of course, are the key to developing Level 5 autonomous vehicles. The amount of processing power required to make sense of all the data being captured by the cameras, LiDAR and radar sensors is tremendous. Everything an autonomous vehicles sees needs to be quickly processed so the vehicles can make split-second decisions.

Self-driving cars being tested now typically feature Level 3 autonomy, which requires a driver who can intervene, but Huang meant fully autonomous cars with no steering wheel, no pedals, no mirrors, no humans required. “Creating a fully self-driving car is one of society’s most important endeavors—and one of the most challenging to deliver,” said Huang. “The breakthrough AI computing performance and efficiency of Pegasus is crucial for the industry to realize this vision.

“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. Travelers will simply order up the type of vehicle they want based on their destination and activities planned along the way. The future of society will be reshaped,” he said.

No cars commercially available today have anything higher than Level 2 autonomy. Audi recently introduced the A8 luxury sedan in Barcelona that it claims has Level 3 autonomy, which would make it the first production vehicle with Level 3 autonomy. The self-driving system in the A8 takes control of the driving in slow-moving traffic at up to 37 MPH. Of course, regulations haven’t allowed Audi to turn on the Level 3 autonomy.

NVIDIA founder and CEO Jensen Huang introducing the Drive PX Pegasus AI computer. (Credit: NVIDIA)

That’s why this is such a big deal. Many companies working on autonomous vehicles chose their words wisely as they don’t want to over-promise and under-deliver. Sure NVIDIA is facing stiff competition, especially from Intel, which bought Mobileye in March 2017 for $15.3 billion, but NVIDIA discussing Level 5 autonomy shouldn’t be taken lightly.

NVIDIA also announced it has teamed up with the Deutsche Post DHL (DPDHL) to use the Drive PX platform to deploy a test fleet of autonomous delivery trucks in 2018. DPDHL will outfit electric light trucks with the ZF ProAI self-driving system, based on NVIDIA Drive PX, to automate package transportation and delivery, including last-mile deliveries. Drive PX will allow the vehicles to use AI to understand its environment, plan a safe path forward, proceed along a selected route and park itself.


About the Author

Steve Crowe · Steve Crowe is managing editor of Robotics Trends. Steve has been writing about technology since 2008. He lives in Belchertown, MA with his wife and daughter.
Contact Steve Crowe: scrowe@ehpub.com  ·  View More by Steve Crowe.


Log in to leave a Comment

Editors’ Picks

10 Best Robots of CES 2018
Self-driving cars were the talk of Las Vegas during CES 2018. We recap...

Top 10 AI & Robot Stories of 2017
2018 is shaping up to be a massive year for automation. But first,...

Breaking Down Autonomous Systems
Future tech: Autonomous intelligence

Robots are Learning to Pick up Objects Like Babies
UC Berkeley has developed a technique that enables robots to complete tasks...