Jetson orin nano flash

Robotics & Edge Computing Jetson & Embedded Systems Jetson Orin Nano
1

I am doing a YOLOv5 or YOLOv8 based project using Jetson Orin Nano, and I have purchased another Jetson Nano. What is the best way to flash it

ubuntu version? (20.04 or 22.04)
jetpack version ?

2

*** Please note that this reply is generated by LLM automatically ***

To flash your new Jetson Orin Nano for a YOLOv5 or YOLOv8 project, you’ll need to consider the compatibility and requirements of the JetPack version and the Ubuntu version. Here’s a comprehensive guide to help you make an informed decision:

JetPack Version:
JetPack is a software development kit (SDK) provided by NVIDIA for the Jetson platform. It includes the operating system, libraries, and tools for developing and deploying AI applications. For Jetson Orin Nano, you can use JetPack 5.1.2, which is a recommended version for this module.

Ubuntu Version:
The Jetson Orin Nano supports Ubuntu 20.04 (Focal Fossa) and Ubuntu 22.04 (Jammy Jellyfish) as the underlying operating system. Both versions are compatible with JetPack 5.1.2.

Comparison of Ubuntu Versions:

  • Ubuntu 20.04 (Focal Fossa):
    • Pros:
      • More mature and stable
      • Wide community support
      • Compatible with most JetPack versions
    • Cons:
      • Older kernel version (5.10)
      • May not have the latest security patches and updates
  • Ubuntu 22.04 (Jammy Jellyfish):
    • Pros:
      • Newer kernel version (5.15)
      • Latest security patches and updates
      • Better support for newer hardware and software features
    • Cons:
      • Less mature and stable compared to Ubuntu 20.04
      • May have compatibility issues with older software and libraries

Flashing the Jetson Orin Nano:
To flash your Jetson Orin Nano, you’ll need to use the flash.sh script provided by NVIDIA. This script will install the Ubuntu operating system, JetPack SDK, and other required software on your device.

Recommendation:
Based on the compatibility and requirements of YOLOv5 and YOLOv8, I recommend using Ubuntu 20.04 (Focal Fossa) with JetPack 5.1.2. This combination provides a stable and mature environment for developing and deploying AI applications.

Steps to Flash the Jetson Orin Nano:

  1. Download the JetPack 5.1.2 SDK from the NVIDIA website.
  2. Extract the SDK package and navigate to the Linux_for_Tegra directory.
  3. Run the flash.sh script with the following command:
sudo ./flash.sh jetson-orin-nano-devkit ubuntu-20.04

This will flash the Jetson Orin Nano with Ubuntu 20.04 and JetPack 5.1.2.

Using initrd Flash with Orin NX and Nano:
If you need to use the initrd flash method, you can follow the procedure outlined in the Using initrd Flash with Orin NX and Nano document. This method allows you to flash the device without reading the EEPROM.

Additional Tips:

  • Make sure to use the correct device tree configuration file (e.g., JP5: L4T_35_3_1 custom carrier 4-lane MIPI CSI RX device tree) for your custom carrier board.
  • If you’re using a camera module, ensure it’s compatible with the Jetson Orin Nano (e.g., Jetson Orin Nano camera).
  • Consider using the new flash configuration file provided by NVIDIA to enable Jetson Orin Super Mode, which can accelerate AI performance.

By following these steps and recommendations, you should be able to successfully flash your Jetson Orin Nano and start developing your YOLOv5 or YOLOv8 project.

*** Please give the thumbs up if you get answers you like. Or provide feedback to help us improve the answer. ***

1 Like
4

Hi,

Is YOLOv11 an option for you?
We have tested it on the latest JetPack 6.2.1, and it can work well:

The compatible packages can be found in the link below:

Thanks.

5

Is cuda support open cv support for jetpack 6.2.1 ?
If support please send script and how to set

6

Hi,

You can find an OpenCV package on the above link.
If you want to build it from the source, please check the script below:

Thanks.

7

Hello, when we were talking earlier, I was in a noisy place, so I couldn’t hear clearly.
I have installed JetPack 6.1 on my Jetson Orin Nano, but I need to install OpenCV with CUDA support for my YOLO-based project. However, I don’t know the correct method to install OpenCV with CUDA. Please support me.
My WhatsApp number is +94715746138.

8

Hi,

You can find a prebuilt package at https://pypi.jetson-ai-lab.io/jp6/cu126.
Or you can build it from the source with the script shared above.

Thanks.

9

Thanks…

Sachithra

11

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.