|The Jetson Nano Development Kit (right) and module (left) are designed to provide AI computing at a price point attractive to makers and DIY enthusiasts. (Image source: Nvidia)|
The latest addition to Nvidia's Jetson family of AI computing platforms is cozying into a space normally occupied by Raspberry Pi and Arduino. But where the most popular single-board computers struggle to bring AI and machine learning applications to the edge, Nvidia's new Jetson Nano promises to do just that, without the need for clever workarounds.
“Jetson Nano makes AI more accessible to everyone — and is supported by the same underlying architecture and software that powers our nation’s supercomputers,” Deepu Talla, vice president and general manager of Autonomous Machines at NVIDIA, said during an announcement at Nvidia's 2019 GPU Technology Conference (GTC). “Bringing AI to the maker movement opens up a whole new world of innovation, inspiring people to create the next big thing.”
Nvidia hopes makers and enthusiasts will embrace the $99 Jetson Nano developer kit in creating projects such as robots, drones, smart devices, and more. According to the company, the Nano platform supports high-resolution sensors and can process sensor inputs in parallel. It can also run multiple neural networks on each sensor streams and supports most of the most popular AI frameworks available today including TensorFlow, PyTorch, Caffe, and MXNet.
What this means for makers is being able to create devices that can handle multiple machine learning tasks such as computer vision and natural language processing, all on a single, compact computer.
For companies looking to build end-use edge computing systems Nvidia is also planning to release a $129 version of the Jetson Nano as well. Nvidia says the Jetson Nano module will help companies reduce development time and achieve a faster time-to-market by reducing the time spent in hardware design, testing, and validation. It will also address size, power, cost, and compute density challenges inherent in developing AI-focused and smart devices.
The Nano module measures 70mm x 45mm, according to company specs, and consists of a 128-core GPU, quad-core Arm A57 CPU, 4 GB of 64-bit LPDDR4 memory, and 16GB of built-in flash storage. It offers gigabit ethernet connectivity, camera inputs, and can support video at up to 4K resolution at 60 frames per second. It comes ready with full Linux operating system support as well. All in, Nvidia says the computer is capable of delivering 472 GFLOPs of performance and consumes only 5-10 watts of power.
The Jetson Developer kit comes in slightly larger at 100mm x 80mm and relies on a microSD card for storage. However it has four USB 3.0 and USB 2.0 Micro-B ports for connecting peripherals and other devices. It also offers the same computing power, memory, and video capabilities.
Earlier generation of the Jetson have been implemented into powerful applications such as drones. And the company offers similar dedicated AI computing platforms for autonomous vehicles. With this latest addition, it's become clear that Nvidia is aiming to become a go-to provider of AI computing platforms for projects at all levels from DIY all the way up to heavy industrial.
The NVIDIA Jetson Nano Developer Kit currently available. The Jetson Nano module will begin shipping in June.
Chris Wiltz is a Senior Editor at Design News covering emerging technologies including AI, VR/AR, blockchain, and robotics.
The nation's largest embedded systems conference is back with a new education program tailored to the needs of today's embedded systems professionals, connecting you to hundreds of software developers, hardware engineers, start-up visionaries, and industry pros across the space. Be inspired through hands-on training and education across five conference tracks. Plus, take part in technical tutorials delivered by top embedded systems professionals. Click here to register today!