Build your Own Autonomous Deep-Learning Robot

Learn deep learning the fun way!  There’s nothing more fun than building a robot and no better way to learn than by doing.  Hardware and software technologies have improved to the point where you really can build an autonomous robot in a few hours.


You will build your own Raspberry PI/TensorFlow powered robot that you can take home and show your friends.  In the process you will get an overview of hardware (everything from soldering to I2C) and software (raspberry PIs, linux, python, tensorflow, deep learning)



This class can support a wide range of skill levels.  The more experience you have, the more customization you will be able to do, but everyone should be able to get the robot running autonomously.

You need to bring a laptop (or a keyboard, mouse and monitor) to program the robot.


Hour 1: Setup the Raspberry PI with Ubuntu.  Install Python, and requisite software/libraries.  Get OpenCV installing and then set aside.

Hour 2: Build the Robot chassis.  Solder the motor hat.  Learn to use a multimeter.  Attach sonars.  Install camera.

Hour 3: Get the Robot moving non-autonomously, powered by an nginx webserver.

Hour 4: Make TensorFlow do object recognition.  Turn on autonomous motion.


I will provide for every student:

  1. Raspberry PI 3
  2. Camera
  3. 16GB SD Card
  4. Adafruit Motor HAT
  5. 2WD Robot Chassis
  6. Small USB Speaker
  7. Sonar
  8. Wires
  9. Laptop Battery

Optional Materials

Bring your own fabrics, wheels, stickers, microphones etc. and you can make your robot even smarter/faster/stylish.