Nvidia Moves Self-Driving Car Tests from Road to Data Center

Nvidia Moves Self-Driving Car Tests from Road to Data Center

The system is based on two servers with one simulating the vehicle's sensors and the other the vehicle's software that processes the simulation data.

The decision comes following the fatal Uber crash in Arizona, which may not have been caused by the self-driving technology, but still yielded some pause from other self-driving vehicle makers.

The technology is created to allow automakers and others to validate their technology on billions of driving miles and increase the strength of their algorithms by repeatedly testing hard scenarios, which would be impractical in the real world. The first server powers the Nvidia DRIVE Sim, a software set that emulates a self-driving car's various sensors, including its cameras, lidar and radar.

"With DRIVE Constellation, we've accomplished that by combining our expertise in visual computing and data centers".

The Nvidia DRIVE platform is now used by over 370 companies developing self-driving technology, including automakers and robotaxi companies and those making self-driving hardware, such as sensors, said Huang.

More news: Heroic police officer Arnaud Beltrame honoured in special Sunday service

Nvidia is temporarily suspending its self-driving auto tests on public roads worldwide. "With virtual simulation, we can increase the robustness of our algorithms by testing billions of miles of custom scenarios and rare corner (edge) cases, all in a fraction of the time and cost it would take to do so on physical roads".

The simulation software generates "photoreal data streams", simulating things like varying weather conditions (rainstorms, snowstorms), blinding glare at different times of day, reduced visibility at nighttime, different terrains and road surfaces. This way companies can test how their cars react to unsafe situations without endangered people or the product.

DRIVE Constellation will be available to early access partners in the third quarter of 2018. More recently, GPU deep learning ignited modern AI - the next era of computing - with the GPU acting as the brain of computers, robots and self-driving cars that can perceive and understand the world.

"With just 10,000 Constellations we can cover 3 billion miles per year", Huang said. Nvidia is now leading the charge in developing technologies for self-driving vehicles, and it's not taking the news of Uber's fatal crash lightly.

Related Articles