C++ Application Template for BeagleBone & Desktop

This project template provides a structured starting point for building and running C++ applications on both your local machine and a remote BeagleBone Black (BBB).

It supports development in Visual Studio Code as well as from the terminal, with an automated build and task system using makefile and manage_cube.sh.

---

📁 Project Structure

BBBBalancingCube/
├── include/        # Header files for application logic
├── src/            # Source files (must match include declarations)
├── lib/            # Local libraries (grouped by domain)
├── .vscode/        # VS Code tasks, debugger configs, settings
├── makefile        # Central build file (local or cross)
├── manage_cube.sh  # Helper script for building/deploying from terminal
├── connect_cube.sh # Helper script to establish and manage connection to remote BBB
├── tools/          # directory for additional tools (calibration, data post processing etc.)

You must follow this structure for the build system and tasks to work properly. Read the file lib/README.md for information on the provided framework.

---

Getting Started

If your machine comes with any of these pre-installed skip the installation (the emsys VM can skip 2.) Get this repo by downloading the zip or cloning it:

git clone git@github.com:ruben01egle/BBBCube.git

1. Install VS Code

• Set up apt-repository following: https://code.visualstudio.com/docs/setup/linux or download from: https://code.visualstudio.com/download (do not use snap)

2. Install Required Tools

Install Make:

sudo apt update
sudo apt install make

Compiler for native/local development:

sudo apt update
sudo apt install build-essential gdb

Cross Compiler for BBB developement:

sudo apt update
sudo apt install g++-arm-linux-gnueabihf

Install autossh to connect to remote server:

sudo apt update
sudo apt install autossh

Optional: Install required python packages in a virtual environment:

./create_venv.sh

If you don't want to use a virtual environment install the required python packages globally:

Install OpenCV for python for the live video stream:

sudo apt update
sudo apt install python3-opencv

Install python QT packages for live data visualization:

sudo apt update
sudo apt install python3-pyqt5
sudo apt install python3-pyqtgraph

3. Install Required Extensions

Search and install the following extensions in VS Code:

• C/C++ (by Microsoft) – for IntelliSense and debugging
• Makefile Tools – for make-based builds
• Task Explorer by Scott Meesseman – to execute tasks from the primary sidebar with one click

4. Open the Project Folder

Open the repository with VS Code.

5. Connect to the cube from remote network

The script connect_cube.sh opens and manages a SSh tunnels to the remote MSys-Server for your group. It establishes port forwarding to all required devices for each group - including the BBB, the servo controlling the cube, webcam and power socket. Start the connection:

./connect_cube.sh start groupX

Then follow the promts in the terminal and keep the connection open. Monitor the connection:

./connect_cube.sh status groupX

Close the connection:

./connect_cube.sh stop groupX

Open a browser and acces the page localhost:42000 to turn on the power for the BBB.

Now you can reach the BBB with SSH:

ssh root@localhost -p48000

To see the live camera feed and change the position of the cube run the python gui cube_control.py.

5. Use Predefined Tasks

• Open the Task Explorer in the VSCode Explorer sidebar and open the vscode tasks.
• Click to run:
  • build native
  • run native
  • build for BBB
  • run on BBB
  • gui control
  • gui viewer
• Debugger are setup:
  • Debug Local
  • Debug on BBB

All configurations (compilers, debuggers, paths) are preconfigured in .vscode/tasks.json and .vscode/launch.json.

6. Set up SSH Key Authentication for Root Login on BeagleBone Black (BBB)

If you start out fresh, it’s a good idea to configure SSH key authentication to avoid having to enter the root password every time you connect to the BeagleBone Black (BBB).

To do this, follow these steps:

1. Ensure you have access to the BBB: Make sure you can still log in to your BeagleBone Black using the root password. If you don't have access, make sure the BBB is properly connected and accessible over the network.

2. Run the SSH key setup script: Use the task copy ssh-key to BBB and enter the root password to copy your PC's ssh-Key to the BeagleBone Black, allowing for password-less login in the future. The manage_cube.sh script also contains a helper function. To copy your SSH key to the BBB, run the following command:

./manage_cube.sh --copy-sshkey

7. Run code on BBB

Use the pre defined tasks in VSCode or the script manage_cube.sh to launch tasks on the BBB. To visualize and record the live data sent form the BBB run the python gui cube_data_viewer.py directly in the terminal, with the pre defined VSCode task or with the script manage_cube.sh. The data gets stored in the folder ExperimentsData/ with a timestamp in a csv-file. You can influence the data visualized and recorded by changing modes in the file gui/cube_data_viewer.py:

plotter = PlotManager(mode="state")
...
recorder = DataRecorder(file_path, mode="imu+state")

---

💻 Terminal-Only Usage (no VS Code)

(warning: debugger is setup in VS Code) You can also use the project entirely from the terminal and any text editor:

Check ./manage_cube -h for more information