# Getting Started

Before You Read

If you haven't already, we'd recommend that you check out the ROS Tutorial Project. This is a tutorial that will allow you to run through all the development basics on a test repo. It also has some Linux tips that you will find useful if you are unfamiliar.

# Cloning the Vehicle Repo

The vehicle repo is where all of the vehicle software is stored. You can clone this directory anywhere, but I'd recommend that you create a separate directory, say soar/, in order to separate all your vehicle files. In this new directory, run:

$> git clone --recurse-submodules https://gitlab.msu.edu/canvas/soar/software/vehicle.git

And that's it for vehicle code.

# Spinning up the Docker

For development purposes, it is ideal to keep user environments uniform so that code works everywhere, not just on whatever system you happen to be on. For this, we use Docker, which is a container that standardizes our OS and dependencies across systems. When you run our Docker container, you are running a complete copy of the vehicle system, just without any sensors or vehicle hardware attached.

Our docker repository is where you will find the source. You should have a workable linux environment if you followed the ROS tutorial project, so spin that up. Firstly, cd into the same directory where you stored vehicle/ and clone the docker repo on commandline with:

$> git clone --recurse-submodules https://gitlab.msu.edu/canvas/soar/software/docker.git

Then cd into the docker/ directory and run:

$> sh build.sh

This will build the docker image. Don't be alarmed if this takes a while, you are building an entire OS from scratch. To run the docker image, run:

$> sh run.sh

And now your docker image is running!

# Building the Repository

Finally, you can build the vehicle code. As our vehicle codebase is a mix of C++ and Python, it does need to be compiled and built before running. Run

$> cd vehicle/
$> catkin_build

For next steps, see Branching