Nvidia & Continental Partner on Xavier-Based Vehicle Systems

Nvidia and automotive technology specialist Continental have teamed up to build an advanced self-driving vehicle systems based on the chipmaker’s Xavier Drive platform.

The collaboration will result in an artificial intelligence (AI) system that will offer support for autonomous vehicles spanning in complexity from Level 2 — basic driver assistance features — to Level 5, where the vehicle has no steering wheel or pedals.

The two companies will focus initially on highly automated driving features, and 360-degree perception and automatic lane changing on highways, including the ability to merge in traffic.

The system will also integrate high-definition maps and provide updates that will enable vehicles to localize themselves.

The Drive platform includes the Xavier system-on-a-chip (SoC), as well as the Nvidia’s Drive operating system and Drive AV (autonomous vehicle) software stacks.

These Nvidia technologies will then integrate a range of Continental sensors, including radar, camera and high-resolution 3D Lidar.

Continental also specializes in advanced driver assistance systems (ADAS), integrating multi-function cameras, fish-eye cameras with surround view and central control units for assisted and automated driving.

“Partnering with Nvidia will enable Continental to advance beyond its leadership position in advanced driver assistance systems, delivering a scalable range of automated and autonomous vehicle systems, from today’s requirements up to Level 5,” Luca De Ambroggi, an analyst with IHS Markit, wrote in a February 5 statement.

The planned market introduction in 2021 aligns with German luxury automaker BMW’s plans to introduce fully self-driving cars to the market that same year.

“The complexity of autonomous driving requires nothing less than the full computational horsepower of an AI supercomputer,” Dr. Elmar Degenhart, CEO of Continental, noted in a statement. “Together with the performance and flexibility of Nvidia’s AI self-driving solution, from the cloud to the car we will achieve new levels of safety, comfort and personalization for future vehicles.”

Xavier is also key part of the Drive Pegasus AI computing platform, a supercomputer that is the size of a license plate, and is designed for fully autonomous Level 5 robotaxis.

Nvidia’s Drive Xavier autonomous machine processors power the Drive software stack, which the company has expanded to a trio of AI platforms covering each aspect of next-generation cars.

With more than 9 billion transistors, Xavier is built around a custom 8-core CPU, a new 512-core Volta graphics processing unit (GPU), a new deep learning accelerator, new computer vision accelerators and new 8K HDR video processors.

“We now have all the key elements in place to take AI self-driving cars from development to mass production,” Jensen Huang, founder and CEO of Nvidia, wrote in a statement.

Nvidia was part of a Mercedes-Benz PR event in Amsterdam earlier this month when the 2019 A-Class was unveiled in the Noord district on the city’s waterfront.

The new A-Class cars are the first to offer the Mercedes-Benz User Experience — or MBUX — infotainment system, which uses Nvidia technology for its natural language processing and augmented reality-enhanced navigation system.

Additional computational capabilities are in reserve, enabling new applications to be deployed over the life of the vehicle, and the car was designed as a learning system, anticipating user preferences the longer the system is deployed.

— Nathan Eddy is a filmmaker and freelance journalist based in Berlin. Follow him on Twitter @dropdeaded209_LR.