LEGO sorter can easily recognize any part of the designer ever created.
An Engineer Daniel West created a sorter for LEGO parts based on artificial intelligence. The versatile LEGO sorter uses a neural network to identify, classify and organize plastic parts more efficiently than a human could. A device was created from more than 10 thousand LEGO parts. Parts are loaded onto a moving tape, through which they fall into sorting, where a camera is installed that takes all the details on the video.
Afterwards, the Raspberry Pi minicomputer processes the video stream and transfers the data to a laptop running the application, which is a convolutional neural network. Artificial intelligence-driven software compares each element with a database containing 3D models of each LEGO element ever created.
As soon as the neural network matches the part with the part number, it sends the data back to the sorter, which at that moment knows which of the 18 sorting baskets to place it in. The machine processes approximately one brick every two seconds.
Such inventions emphasize the flexibility of LEGO as a building platform. Young children can stack bricks to create simple objects, while older children and adults can develop amazingly efficient hardware by bringing technology to the designer.
Amazon and LEGO are currently running a contest that actually integrates LEGO with technology. Both companies have instructed builders to integrate Alexa voice commands with LEGO Mindstorm kits and will reward the best creations with a trip to Denmark.
RECOMMENDED TO READ: