NVIDIA Announces Availability of Jetson AGX Xavier Production Modules

Jetson AGX Xavier Module. Source: NVIDIA

December 12, 2018      

NVIDIA today announced general availability of production modules for its Jetson AGX Xavier platform, aimed specifically at robotics developers building the next generation of autonomous machines. The modules, which cost $1,099 in quantities of 1,000 or more, are now available from distributors worldwide, including Arrow Electronics, NVIDIA said.

In September, the company announced the Jetson AGX family at its GTC Japan Event. Today’s announcement was made at a developer’s event in Santa Clara, where more than 400 developers met to learn more about the platform.

Higher operations needed for autonomous systems

The system includes a small computer module capable of up to 32 trillion operations per second (TOPS), with multiple operating modes (10W, 15W, and 30W). In addition, the AGX Xavier platform is more energy efficient than its predecessors, including the Jetson TX1 and TX2. The company also announced the newly available TX2 4GB model, with a lower-cost model than the 8GB version.

Jetson AGX Xavier article

NVIDIA’s Jetson family of embedded AI processors. Source: NVIDIA

In addition to the module, NVIDIA provides support and tools for developers, including the Jetpack and DeepStream software development kits. The SDK for autonomous machines includes support for AI, computer vision, and multimedia, among other features.

With its smaller footprint and power efficiency, NVIDIA said Jetson AGX Xavier is “ideal for deploying advanced AI and computer vision to the edge, enabling robotic platforms in the field with workstation-level performance and the ability to operate fully autonomously without relying on human intervention and cloud connectivity.”

Applications that could benefit from this edge technology include package delivery and industrial inspection systems, which would require advanced levels of real-time perception and inferencing to perform, NVIDIA said.

The high-performance system “can handle visual odometry, sensor fusion, localization and mapping, obstacle detection, and path-planning algorithms critical to next-generation robots,” the company added.

Major users of Jetson AGX Xavier

Since making the developer’s version available in September, several companies have begun work on building systems with the new platform, NVIDIA said. Oxford Nanopore, a U.K. medical startup, said it is using Jetson AGX Xavier for its MinIT handheld AI supercomputer to perform real-time analysis with its MinION handheld DNA sequencer.

“MinIT can be nearly 10 times more powerful than standard laptops and brings portable, real-time sequencing to more people in more locations,” said Gordon Sanghera, CEO of Oxford Nanopore.

In the robotics space, Japan’s Denso is also using the platform to achieve more efficiency in the global auto parts maker’s operations.

“Denso can leverage its long history in auto parts manufacturing to bring AI to factories, boosting productivity and efficiency while increasing workplace safety,” said Katsuhiko Sugito, executive director at Denso. “We believe that Jetson AGX Xavier will be the key platform driving this initiative.”

Additional companies that announced working with Jetson AGX Xavier include Yamaha, Komatsu, JD.com, Meituan, FANUC, Kawada, and Musashi Engineering.