OSR-course

A repository to simplify the environment setup for a open-source-robotics course

This course is built on Ubuntu16.04 with ROS. For now it only works with nvidia gpu based host.

Basic setup: Docker, Nvidia-docker, course content

To start with, make sure you installed docker. Installation instructions can be found at install docker.

Then, install nvidia-docker such that docker run command recognizes --gpus all tag. You can find out more at NVIDIA/nvidia-docker.

Setup you working directory

git clone https://github.com/HoraN1/osr-course.git

To get the course content, in current path, run

bash get-course.sh

And you will see a new directory called catkin_ws in you current directory.

To build the osr-course image and run the image

Build with default name crigroup/osr-course

docker build -t crigroup/osr-course .

Or if you want to customize yourself

docker build -t <name:tag> .

But remember to modify the image name and tag in the last line of docker-run.sh. And to run a container from the image, run

bash docker-run.sh

You can change the container name with the tag --name=<CUSTOM_NAME> in docker-run.sh file.

To test if the X-server is forwarding

In the docker container, run:

bash test-gui.sh