Skip to main content

Mini Project: Humanoid Nervous System

Goal​

In this mini-project, we will integrate all the ROS 2 components and concepts learned in this module to build a basic "nervous system" for our humanoid robot. This includes:

  • Running talker and listener nodes.
  • Visualizing the humanoid.urdf in RViz2.
  • Operating the control_loop node with dummy sensor input.
  • Sending commands and receiving data through the ai_bridge node.

Integration Steps​

  1. Start the ROS 2 environment:

    cd ~/ros2_ws
    colcon build --symlink-install
    source install/setup.bash

  2. Launch visualization and control:

    # Open three separate terminals
    # Terminal 1: Launch RViz2 and robot_state_publisher
    ros2 launch ros2_foundations visualize.launch.py

    # Terminal 2: Run the control loop node, which also publishes dummy IMU data
    ros2 run ros2_foundations control_loop

    # Terminal 3: Run the AI bridge node
    ros2 run ros2_foundations ai_bridge

  3. Interact with the system:

    • Observe the robot model in RViz2: The shoulder_joint of the humanoid will articulate based on the simulated IMU data from the control_loop node.
    • Monitor ai_bridge output: The bridge node will log when it receives IMU data and attempts to send it to an "external AI".
    • Test AI commands (optional, requires a separate AI client node publishing to /ai/command): 
      ros2 topic pub /ai/command std_msgs/String "data: '{\"action\": \"move_head\", \"direction\": \"left\"}'" --once
      The ai_bridge node will log this command.

Conclusion​

This mini-project demonstrates the foundational components of a robot's nervous system using ROS 2, from basic communication to sensor processing and AI integration.