Nvidia has revealed its latest computer chip named 'Xavier' that is the size of a postage stamp yet capable of performing 320 trillion operations a second, and will likely form the basis of a Level 5 autonomous driving system to underpin future driverless cars.
The US-based technology company claims the piece of hardware is the world’s most complex computer chip and could revolutionise transport as the brain for millions of autonomous cars set to hit the road in coming years.
Opening the 2018 Consumer Electronics Show in Las Vegas with a presentation featuring Volkswagen and Uber – customers that will use its hardware to power next-generation autonomous vehicles - Nvidia chief executive Jensen Huang spoke of how 2000 engineers worked for four years with a $US2 billion budget to develop “Xavier” a tiny chip with 8K video processors, an 8-core CPU and 512-core GPU, deep learning ability and more. Working with other hardware in a numberplate-sized package known as “Pegasus”, the result, Huang says, is an automotive brain capable of conducting 320 trillion operations per second.
That could serve as the brain of completely driverless “level five” autonomous cars offered by Volkswagen, Uber and 23 other companies relying on Nvidia tech to develop next-generation transport.
Huang says the systems represent some of the most complex computing projects in history.
“Now, we’ve built PCs, laptops, game consoles, supercomputers, and I can tell you, without exception, building a computer for autonomous vehicles is of a level of complexity the world has never known,” he says.
“This computer is on all the time, monitoring all of the sensors that are coming at it — it can never fail. It can never fail because lives are at stake.
“And it has to make the right decision, running software the world has never known how to write.”
Though Nvidia will work with German maker Volkswagen on its reborn Kombi van based on the popular I.D. Buzz concept the partnership will go beyond self-driving systems and develop technology that can unlock and start cars using facial recognition features as opposed to keys, track a driver’s sight lines to make sure they are paying attention, use normal conversational language to access key features, warn drivers of hazards using augmented reality displays and work with natural gestures to interact with key systems.
“In just a few years, every new vehicle should have AI assistants for voice, gesture and facial recognition as well as augmented reality,” Huang says.
“Working with Volkswagen, we are creating a new generation of cars that are safer, more enjoyable to ride in than anything that has come before, and accessible to everyone.”
Volkswagen chief executive Herbert Diess says “artificial intelligence is revolutionizing the car”.
“Autonomous driving, zero tailpipe emission mobility and digital networking are virtually impossible without advances in AI and deep learning,” he says.
“Combining the imagination of Volkswagen with NVIDIA, the leader in AI technology, enables us to take a big step into the future.”