Teaching Self-Driving Cars to Drive
NVIDIA Corp. has been making interactive graphics processors for years. Now it’s working with more than 80 carmakers and suppliers to bring fully autonomous vehicles to market.
NVIDIA Corp. has been making interactive graphics processors for years. Now it’s working with more than 80 carmakers and suppliers to bring fully autonomous vehicles to market.
Danny Shapiro, senior director, Automotive, says the NVIDA Drive PX 2 system is a supercomputer for self-driving cars that collects and processes continuous streams of data from an autonomous vehicle’s sensors to detect its surroundings. The system can process 24 trillion operations per second. The result is what Shapiro describes as superhuman levels of perception.
The system also uses artificial intelligence to quickly “learn” safe driving techniques from simulations, millions of miles of on-the-road videos and input from professional drivers.
RELATED CONTENT
-
On Fuel Cells, Battery Enclosures, and Lucid Air
A skateboard for fuel cells, building a better battery enclosure, what ADAS does, a big engine for boats, the curious case of lean production, what drivers think, and why Lucid is remarkable
-
TRW Multi-Axis Acceleration Sensors Developed
Admittedly, this appears to be nothing more than a plastic molded part with an inserted bolt-shaped metal component.
-
On Automotive: An All Electric Edition
A look at electric vehicle-related developments, from new products to recycling old batteries.