Isaac_ros_nitros colcon build fails in multiple ways on docker image/container

Where can I find flat_scan.hpp that is in several *.cpp and *.hpp files so I can colcon build isaac_ros_nitros and other isaac_ros repos?

That file is not in NVIDIA Isaac ROS · GitHub nor in nor on Jetson AGX Orin 32gb developer kit.

If I substitute nitros_flat_scan.hpp in the .cpp or .hpp it really does an interesting fail warning/error output.

And, finally, perhaps I’m old but how do you use a docker container to experiment with ros2 / isaac when the container goes away when you exit? Or should I remove --rm from or docker commit every time I might need to shutdown or reboot?

If there is a somewhere that explains that please just point me to it.


It seems like you may be referring to isaac_point_cloud_interfaces/msg/flat_scan.hpp, which is a file that is automatically produced by the ROS IDL C++ generator when the package is built. The generator converts the ROS language-agnostic interface definitions (flat_scan.msg) into language-specific interfaces (flat_scan.hpp).

The Isaac ROS Docker container and associated script provided in the Isaac ROS Common repository are tools that help streamline the process of installing the dependencies necessary for Isaac ROS and ROS 2 more broadly on x86 and Jetson platforms. All of the source code for your ROS workspace is mounted from the host computer’s filesystem into the container, which means that changes you make to your ROS code will persist even after the container has been shut down. If there are additional directories that you would like to persist after a container shutdown, you can use the Docker -v flag to bind the appropriate volume to an absolute path inside the container. The Docker reference is a good resource for more details.

Thank you for replying. The information you provide is helpful and helps resolve my problem. The only thing is I guess on docker a person just never updates anything because it is deleted whenever the container is rm and the image remains pristine.

1 Like