Building A.I. Controlled LEGO Robots
Updated: Feb 9, 2021
LEGO forms an integral part of many people's childhood. Although LEGO bricks have gone through a series of evolutionary changes since their inception in 1949, the purpose of providing a system for creative play has remained the same. The development of LEGO Mindstorms to give people the ability to construct programmable robots from LEGO building blocks is arguably one of the most significant revolutionary changes in its history.
Since the original Robotic Invention System introduction in 1998, the NXT, the NXT 2.0, the EV3, and the Robot Inventor kit have gradually expanded their motor and sensor capabilities. Even though these incremental improvements are essential for the LEGO Mindstorms platform's maturation, everyone is still holding their breath for the next major revolution.
As a LEGO enthusiast, I want to define the next LEGO revolution rather than waiting for someone else to invent it. The most important revolution in the technology world is the advancement of Artificial Intelligence (A.I.) in recent years. We are surrounded by these incredible A.I. systems in our daily lives, from the smart assistant on your phone to the increasing number of self-driving capable cars on the street. It naturally follows that the next major LEGO revolution needs involve A.I. This is exactly what a few like-minded friends and I set out to do.
We are pleased to introduce the Cortic A.I. Toolkit (CAIT) to our fellow LEGO builders. You can experience this LEGO revolution today, and it is FREE for everyone. Here is a quick rundown of what CAIT can do for you and your LEGO projects.
Instantly add Computer Vision, Speech Recognition, Natural Language Processing, Speech Generation, and Smart Home Control capabilities to your existing LEGO Mindstorms EV3 or Robot Inventor kit creations.
The ability to mix and control up to 8 LEGO hubs (EV3 or Robot Inventor) wirelessly (using Bluetooth) using a single $35 Raspberry Pi 4B.
Visual programming capability utilizing an extension of Google's Blockly code editor
Robust Python API to support all the above features
Tools to convert your visual programs into Jupyter Notebook format or straight Python code for easy data visualization and code experimentation.
It's effortless to get started. Just clone the CAIT project from Github and follow our Quick Start guide to get the host system up and running. This short setup guide presents a step-by-step process to connect both the EV3 and Robot Inventor kit hubs to the host system via Bluetooth. That's it. You should now be able to experiment and add various A.I. capabilities to your LEGO projects in minutes.
We are confident that you will enjoy using CAIT. We would love to hear about your A.I. powered LEGO builds on our project's Discussions page. We hope this project will be able to stand the test of time and live up to the LEGO group's motto, "only the best is good enough."