`acceleration_firmware_jetson_nano`

Board	Picture	Description
Jetson Nano Developer Kit		NVIDIA® Jetson Nano™ Developer Kit is a small, powerful computer that lets you run multiple neural networks in parallel for applications like image classification, object detection, segmentation, and speech processing. All in an easy-to-use platform that runs in as little as 5 watts.

This repository provides firmware artifacts for the NVIDIA Jetson Nano board meant to accelerate ROS 2 robotic applications.

NOTE: The binaries in this repo take various GB of data. Due to GitHub size restrictions, the whole source code is available in the following links:

acceleration_firmware_jetson_nano_0.1.0.zip (3 GB)

Hardware Acceleration capabilities

According to REP-2008's proposal.

Capability	`0.1.0`
`1.` Kernel Levels
`1.i` level I kernels	⚠️ (only if VPI-enabled, see README)
`1.ii` level II kernels
`1.iii` level III kernels
`2.` Build System
`2.i` ament extensions	✓
`2.ii` `ament_acceleration` support
`3`. Build Tools
`3.i` hardware emulation (`hw_emu`)
`3.ii` hardware emulation (`sw_emu`)
`3.iii` image tooling
`3.iv` Linux kernel
`3.iv.a` modern Linux kernel
`3.iv.b` LTS Linux kernel
`3.v` hypervisor
`3.v.a` no control domain VMs
`3.v.b` guest VMs in disk
`3.v.c` control domain in disk
`3.v.d` no control domain VMs in disk
`3.vi` network booting
`3.vi.a` boot artifacts
`3.vi.b` rootfs
`3.vi.c` multi-network boot
`3.vi.d` secure network booting
`3.vi.e` save in disk network boot
`4.` Benchmarking
`4.i` kernel benchmarking	✓
`4.ii` ROS 2 acceleration benchmarking	✓
`5.` Documentation
`5.i` in-code documentation
`6.` Testing and CI
`6.i` `acceleration_examples`	⚠️ (see https://github.com/ros-acceleration/isaac_ros_image_pipeline)