NVIDIA announced the GTC 2019 keynote speech in Silicon Valley. NVIDIA gave a speech and officially released the robot developer toolbox Jetson Nano, which can introduce AI computing power for robot design. The Jetson Nano is an embedded computing device similar to the Raspberry Pi. It is equipped with a quad-core Cortex-A57 processor. The GPU is an NVIDIA Maxwell architecture graphics card with 128 NVIDIA CUDA cores. The memory is 4GB LPDDR4 and the storage is 16GB. eMMC 5.1 supports 4K 60Hz video decoding.
NVIDIA said that the Jetson Nano can provide 472 GFLOPs for running modern AI algorithms fast while powering consumption is just 5 to 10 watts. In addition, the Jetson Nano is very mini, with a length and width of the only 69.6x45mm, making it ideal for use in robots or smart speakers.
The Jetson Nano suite runs directly on Linux and supports a large number of AI frameworks. As for the price, the Jetson Nano is priced at $99. In addition, NVIDIA also offers a “production-ready” version, priced at $129.