2-Wheeled Balancing Robot

Mana Bot

Work has began with the development of a 2-wheel balancing mobile robotics platform. The goal is to develop a versatile home robotic system using open and highly available parts.

The platform in its most basic form is comprised of:

  • Arduino compatible boards
  • An IMU (Inertial Measurement Unit)
  • A multicore computer for driving the architecture
  • A high-power motor driver board
  • 2 gear motors with encoders
  • Plenty of plexiglass sheeting, aluminum trim, and nuts and bolts

There are some things that have been added in order to simplify the experimentation with various facets of robotics and human interaction:

  • Audio input/output:
    • "Auduino" synthesizer for noise-making
    • GinSing, Arduino speech synthesis shield
    • EasyVR, Arduino speech recognition shield
    • Software-based speech synthesizer and recognition applications (identifies robot by name)
  • Computer vision:
    • Dual USB cameras
    • ASUS Xtion
  • Networking:
    • Utilizes WiFi on main board
    • LAN setup over 10/100/1000Base-T
    • Main board assigns DHCP addresses and routes traffic over WiFi
    • RFM22 wireless modules
  • Navigation:
    • IR and sonic rangers for obstacle and drop-off detection
    • Computer vision using opencv
    • Kinect libraries for creating and navigating an internal SLAM map of the environment
  • AI components:
    • ROS "Robot Operating System"

The intent is to make this platform "pluggable," so to speak. If you don't intend to build a 2-wheel balancing robot, you can easily modify a bit of the Arduino code, interface it with a different controller board and motors, and still be able to interface with the rest of the system just fine.

You can view photos here of the initial build.

November 2013 — UPDATE: This project has been abandoned. If you would like to learn more, feel free to contact me and inquire about it.

Bobby Larson

Read more posts by this author.

Subscribe to This is Mana

Get the latest posts delivered right to your inbox.

or subscribe via RSS with Feedly!