Our team at NVidia has created Dockerfile for deploying ROS and ROS2 distributions with the latest version of Jetpack on the Nvidia Jetson platform. The following distributions are supported:
- ROS Melodic (
ros:melodic-ros-base-l4t-r32.4.3) - ROS Noetic (
ros:noetic-ros-base-l4t-r32.4.3) - ROS2 Eloquent (
ros:eloquent-ros-base-l4t-r32.4.3) - ROS2 Foxy (
ros:foxy-ros-base-l4t-r32.4.3)
The dockerfile are available here: https://github.com/dusty-nv/jetson-containers
We have also update the ROS Deeplearning repo (that offers implementation for object detection, classification and segmentation using over 40 different pre-trained models) to support ROS2 and the following input sources:
- MIPI CSI cameras
- V4L2 cameras
- RTP / RTSP
- Videos & Images
- Image sequences
More details are here: https://github.com/dusty-nv/ros_deep_learning
If you face any problem please open an issue on the github repo and we will try to address them soon.