Local simple environment

This dockerfile aims to provide with only the most essential tools to work with the vision pipeline of the autonomous vehicle, namely Ros2, Python 3 (and its most used libraries) and GTSAM. It is intended to be more accessible since it doesn't require complicated configuration of computer nor specific hardware, but if you do have Nvidia graphics card, then is it better to use a different version that uses CUDA to achive better performance.

The environment does NOT have Zed tools.

Requirements

• Docker installed and running
• Minimum 15Gb free storage space
• Mininum 8GB RAM

Setup

Note

For Windows users, you may need to open a terminal and enter WSL to be able to run the following commands.

1. Build image

To build the image from the dockerfile, open a terminal at the current directory and run the following command:

docker build -t <name_you_want_to_give_to_the_image> .

for example, you may write docker build -t tfc-simple-env . (don't forget the dot at the end) Building the image will take a while depending on the performance of your computer.

2. Run container

After building the image, you are ready to enter the container and use the tools inside, to do it run this command:

docker run -it --rm -v <path_to_the_code_repo>:/local-ros2 <name_you_gave_to_the_image>

If the applications in the container need to open a new window, you should add docker to the allowed hosts of xhost in order to allow display forwarding.

xhost +local:docker
docker run -it --rm \
-e DISPLAY=$DISPLAY \
-v <path_to_the_code_repo>:/local-ros2 \
-v /tmp/.X11-unix:/tmp/.X11-unix:rw \
<name_you_gave_to_the_image>
xhost -local:docker # disable display forwarding after running docker