Skip to main content

Machine Learning Meets MINDSTORMS


Machine Learning Extension
LEGO MINDSTORMS announced today a brand new feature that is sure to excite all users. Machine Learning Extension is coming to the software in August 2022. This will allow users to implement new applications that were not possible before. Machine Learning has become exceedingly popular in recent years with applications in all sorts of industries from robotics to financial services to climate change. Now, even the youngest of robot enthusiasts and programmers have access to the world of Machine Learning and Artificial Intelligence.

Comparing two classes
Machine Learning on MINDSTORMS will work with both images and audio. For instance, you could train your model to recognize smiles and frowns on your face and have the emotion displayed on your hub. You could train your model to recognize when a red ball is present or not or identify a particular LEGO element. You can control your robot with audio commands that use your own voice. It is exciting to see what new possibilities the Machine Learning Extension in MINDSTORMS opens up.

Up to six classes can be trained at one time and a minimum of two are needed for the model to work. It is very easy to set up these classes and collect samples. All the steps, including the training, are conducted within the App. Visual probability bars allow the user to see exactly what is going on and determine if they need to make changes.

Using Machine Learning to Identify Sad and Happy Faces

Adding Samples

Dan Meehan, Creative Design Lead said, "Being able to quickly create a Machine Learning model that makes their robot behave differently when it detects their face, or their friends, gives them the confidence to explore, understand and use machine learning in the real world. And if things don't work, they're quick to understand that maybe the model needs re-training with more samples, or they need to reposition the camera to get a better view."

By adding Machine Learning to MINDSTORMS, you can now build even smarter robots and continue your "Learning through Play" journey.

Popular posts from this blog

Celebrating MINDSTORMS with a Remix - Part 3

The ROBOTMAK3RS continued their celebration of the 25th Anniversary of MINDSTORMS through these Fall and Winter remix projects. Each ROBOTMAK3R was tasked with selecting one LEGO set of their choice and combining it with a MINDSTORMS set. Below are the five amazing models they came up with. Braill3 by Jerry Nicholls Braill3 is an EV3-based LEGO Braille bricks reader. This robot uses its fingertip, made from three touch switches, to read messages written using the LEGO Braille bricks and will speak out what it detected. If it sees a simple maths problem it will attempt to solve it and give the answer as well. To learn more about the process of creating this machine, read Jerry's blog . Braill3 can be viewed here . Set Review: The Braille Bricks set is well thought out. The ratios of the letters is suitable for general use and the addition of some punctuation and arithmetic operators is excellent. There is a card showing what bricks there are and their quantities, but no form of sort...

Celebrating MINDSTORMS with a Remix - Part 2

The ROBOTMAK3RS continued their celebration of the 25th Anniversary of MINDSTORMS through these summer and fall remix projects. Each ROBOTMAK3R was tasked with selecting one LEGO set of their choice and combining it with a MINDSTORMS set. Below are the five amazing models they came up with. Remote controlled material handle r by Jozua van Ravenhorst (aka Mr Jo) This remix combines the LEGO Technic Material Handler (42144) with MINDSTORMS EV3 (31313) It uses the power of pneumatic cylinders to move objects around. By using a bluetooth remote control, very precise movements can be made with this model. Touch sensors in the base chassis prevent the turret twisting the cables that go through the turntable to much. The program has several protections to prevent over pressurizing the system for each of the 3 individual pumps and valves that control the 2 booms and claws. The real version of this machine is mostly used in waste material sites to bring the material to machines that sort and...

Celebrating 25 Years of MINDSTORMS

In celebration of the 25th Anniversary of MINDSTORMS, we take a trip through history. Please also visit ROBOTMAK3RS Community every week as we highlight different projects all through 2023 in celebration of the anniversary. Some of the early history is based on the content shared by  Coder Shah  in our  MINDSTORMS EV3 Community Group . Some of the text and links may have been edited from his original posts for consistency and clarity.  1984 - Kjeld Kirk Kristiansen watched a TV program called "Talking Turtle," where MIT professor Seymour Papert demonstrated how children could control robot "turtles" using LOGO, a programming language he developed. 1988 - The collaboration between MIT and LEGO resulted in LEGO TC Logo in 1988, which allowed students to control LEGO models using computer commands. The video shows Papert demonstrating TC Logo. 1990 - LEGO TC Logo was hampered since the robots you built had to be tethered to a personal computer. LEGO and MIT...