Ideal Jetson Platform for ROS and GStreamer Option

Hi,
I have used Jetson Nano before.
My Question = Which is the ideal platform to goto (Xavier NX) or (AGX Xavier), if I am aiming to move in the direction of:

  1. ROS
  2. Gstreamer
  3. Deep Learning
    Please suggest this from the standpoint of currently available and compatible:
  4. Sensors (cheaper options) - especially cameras, lidars, etc.
  5. Device drivers related to the above.

My current opinion:
I thought that AGX Xavier would be suitable, but then found it might have trouble connecting with usual usb cameras, like the ones on Jetson nano? Correct me if I am wrong.

Hi nikhil.saxena,

The major difference between AGX Xavier and Xavier NX are the system performance.
If needs more operations at the same time, then AGX Xavier is the best choice.

I’m not really sure what you mean regarding the usual usb cameras, you may refer to below to find the details:
https://developer.nvidia.com/embedded/jetson-agx-xavier-developer-kit
https://developer.nvidia.com/embedded/jetson-nano-developer-kit

Lets say I want to make a robot using AGX Xavier. What compatible sensors for this platform exist currently in the market and at what minimal cost?

To be specific, which sensor(s) would you recommend for camera, lidar, etc. compatible with AGX Xavier and used for applications related to ROS (Robotic OS) and GStreamer?
I am not sure whether all sensors supported on Jetson Nano, are supported on AGX Xavier? Correct me if I am wrong.

Please find Jetson Partner Support Cameras list from https://developer.nvidia.com/embedded/jetson-partner-supported-cameras

Does AGX Xavier platform works with Intel Realsense Cameras/Lidar?

Or Azure Kinect DK ?

To use what we suggested is more comfortable, but still can use other by your own.
You can search the forum to find similar cases.

Thanks