The nVidia Jetson Nano (™ ® © etc…) is the smallest entry in nVidia’s family of SoM (System on Module) designed for machine learning applications. It takes the shape of a double-height SODIMM and, as SoM usually do, it contains a complete computer system except for the trivial stuff : power supply and connectors.
It’s also remarkably affordable considering its specifications. I think it’s fair to call it the “Raspberry Pi of AI”. In fact, the development board for this SoM has the same GPIO connector as a Raspberry Pi… and I intend to make full use of that.
To whet your appetite, here’s a photo of my Jetson Nano dev kit attached to a board I designed for the Raspberry Pi Zero. It features a Xilinx Artix-7 FPGA and an STM32H750 microcontroller. That’s gonna be a lot of fun for the whole family :
Click image for high resolution
In this section I’ll deal specifically with the Jetson Nano. For all things AI, there is (or will be) a dedicated section on this site, because I don’t plan on getting stuck to just one hardware platform. The expansion board I’ve just teased will also be discussed in its own section. If it ain’t there yet, be patient : this is a marathon, not a sprint.