Development workflow

At this point you should have already setup your RLL workspace and followed the steps from the Getting started guide.

The instructions in the document will help you build and run your project. The commands which are described here can be applied to any project that is available with the RLL. However, we will use the Robot Playground project as an example project and you need to replace the project name in the project specific commands to adapt it to different projects. The project specific commands are all of the format: roslaunch <project-name> <arguments>

General notes

The following commands need to be execute in a terminal window, i.e. copy and paste or type them in the terminal prompt and press Enter to execute them. While a command is running you will see the command output and cannot execute any new commands in this terminal until the command is finished or you stop it.

  1. Any command you should execute will be displayed like this:
<command> <argument1> <argument2>

2. Commands need to be run in a terminal where the ROS environment has been initialized. To initialize the the ROS environment execute the following command:

source ~/rll_ws/devel/setup.bash

You need to do this with every new terminal you open! To avoid having to do this every time you start a terminal it is highly recommended to add the above command at the end of your ~/.bashrc file.


If you are using the RLL VM the ~/.bashrc file has already been adjusted for you and you do not need to do this.

  1. Commands should be run from within your catkin workspace directory ~/rll_ws. To navigate in to this directory execute:

    cd ~/rll_ws

    If you use a different workspace folder other than ~/rll_ws, adjust the path in the above command accordingly.

Build the project

Make sure you are in your catkin workspace directory and your ROS environment is initialized. Run the following command to build the project:

catkin build rll_robot_playground_project

If you are going to write your code in Python you only need to build the project once. If you are using C++ you will need to run this command every time you make changes to your code.

Run the project

During development you run your code in the RLL simulation environment. Therefore, you first need to start simulation and then run your code. Execute each command in a new terminal tab or window.

  1. First, Start the simulation environment:

    roslaunch rll_robot_playground_project setup_simulation.launch

    The terminal will display some log output and will continue to print logging information while you run your code. Additionally, RViz is opened showing the robot visualization. You should already be able to see the robot move into its home position.

  2. Second, execute your project code. Depending on whether you want to use Python or C++ run one of the following commands:

    • Python version (the default):

      roslaunch rll_robot_playground_project run_your_code.launch
    • C++ version:

      roslaunch rll_robot_playground_project run_your_code.launch use_python:=false

    The robot should now begin to move and the log output of the program is printed to the terminal.

To execute the project code again, interrupt the second command you executed and simply re-run it. To interrupt the running command press Ctrl + C in the corresponding terminal.


If something goes wrong or you close a terminal by accident, it is best to close all terminals that are still open. Once all windows are closed repeat all the steps above.

Write your own code

Once you are able to start the simulation and execute the project code you can go ahead and start to modify the project and write your own code.

Usually the project code is included in two versions, one version written in Python and one in C++. The C++ code can be found in the src/ folder whereas the Python code is located in the scripts/ directory within the project folder.

The full path to the project folder is: ~/rll_ws/src/<project-name> I.e. for the Robot playground project it is ~/rll_ws/src/rll_robot_playground_project.

Where to put your code

In case of the Robot playground project the code in the file scripts/ or src/playground.cpp, respectively, is executed. Therefore place your own code in these files.

Depending on the project there will be multiple files, in case of the Robot playground project there are the files scripts/ and src/hello_world.cpp, respectively. These files contain the same code as the project’s main code file. This way you can initially run the project without having to change anything. Feel free to delete and modify the code. You can always use the hello_world file as a reference or to restore its original content.

Re-Build your code

As stated above, if you are using C++ you need to re-build your code every time you made changes to the code and want to test them. Therefore, once again run:

catkin build rll_robot_playground_project

Run your code

You can run your code by following the steps from “Run the project” as shown above. To run your modified code you need to restart the command from step 2. To do this, interrupt the running command by pressing Ctrl + C in the terminal in which you executed the command from step 2. Once the command has stopped simply execute the command again.


If you are unsure about this it is best to close all terminal windows and repeat the steps 1 and 2.

Use C++ by default

If you are programming in C++ you can and should adjust the launch files to use C++ by default.

  1. Open the file launch/move_sender.launch, which is located in the project folder and change the following line, while keeping the remaining lines as they are, from:

    <arg name="use_python" default="true" />


    <arg name="use_python" default="false" />
  2. Repeat step 1 but this time modify the file launch/run_your_code.launch

If you run your code now as described in Run the project, the file playground.cpp is executed by default.


You will need to make this change if you want to submit your C++ code! On the real robot the launch file is executed without overriding the default.

Submit your code

To run your code on a real robot, follow the steps as described in 4. Configure your API access. That is run the following command:

roslaunch rll_robot_playground_project submit_project.launch