We present BricksRL, a platform designed to democratize access to robotics for reinforcement learning research and education.

We present BricksRL, a platform designed to democratize access to robotics for reinforcement learning research and education. The integration of TorchRL with the LEGO hubs, via Bluetooth bidirectional communication, enables state-of-the-art reinforcement learning training on GPUs for a wide variety of LEGO builds. This offers a flexible and cost-efficient approach for scaling and also provides a robust infrastructure for robot-environment-algorithm communication. We present various experiments across tasks and robot configurations, providing built plans and training results. Furthermore, we demonstrate that inexpensive LEGO robots can be trained end-to-end in the real world to achieve simple tasks, with training times typically under 120 minutes on a normal laptop.

Teenager invents robot to solve Rubik's Cube BBCRuarcc the year 10 student who has programmed a robot that can solve a Rubik's Cube puzzle A 13-year-old schoolboy has invented a Lego robot that can solve a Rubik's cube. Ruarcc, from St Malachy's College in north Belfast, first took steps to create puzzle-solving robot prototypes in his second year at school, aged 12. This was made possible after the school launched its creative digital technology hub (CDTH) last year. Teacher Clare McGrath commented she "didn't believe" that Ruarcc's robot would work at first.'People are amazed my robot can solve Rubik's Cube' Ruarcc told BBC News NI it was "frustrating", but he worked on making it better. "People tend to be amazed that it can solve one," he said.

A robot made of Lego can quickly perform an important step for creating machines made of DNA. "This started as a final project in an undergraduate lab course," says Rizal Hariadi at Arizona State University, who tasked his class with building tools using "frugal science". The robot that one group of students built has proved particularly useful and resembles a single arm topped with a holder for cylindrical tubes. It performs a procedure to mix the liquid contents of the tubes, first tilting the tubes from vertical to horizontal, then rapidly spinning them around. This creates a single liquid with a density that uniformly decreases from the bottom to the top.

In the winter of 1997 Carver Mead lectured on an unusual topic for a computer scientist: the nervous systems of animals, such as the humble fly. Mead, a researcher at the California Institute of Technology, described his earlier idea for an electronic problem-solving system inspired by nerve cells, a technique he had dubbed "neuromorphic" computing. A quarter-century later, researchers have designed a carbon-based neuromorphic computing device--essentially an organic robot brain--that can learn to navigate a maze. A neuromorphic chip memorizes information similarly to the way an animal does. When a brain learns something new, a group of its neurons rearrange their connections so they can communicate more quickly and easily.

With today's proliferation of tablets it can be hard to distinguish one device from the next. But here's something that sets the Kno Textbook Tablet apart (besides its big dual displays and focus on students): it's getting stress-tested by a Lego robot. The Kno product development team needed a way to automate tests of the ambient light sensor and the note-taking stylus' interaction with the LCD touch screen. So they built a Lego robotic arm (they give it the far less sexy name of "accelerated life test apparatus") to shoulder the repetitive work. In the behind-the-scenes video below, you'll see the arm directing the pen back and forth and up and down across a screen, while another robot makes the device itself go back and forth.

We've seen Lego robots that do Sudoku puzzles, solve Rubik's Cube, and mimic the human hand. Well, this Lego robot builds Lego models--so far. How long until it starts building other robots out of Lego? Hobbyist Will Gorman built his Lego Mindstorms NXT MakerLegoBot for the ongoing Lego World 2010 expo in Zwolle, the Netherlands. Inspired by the open-source 3D printer MakerBot, the machine is made of more than 2,400 bricks and runs on three NXT Intelligent Bricks and nine NXT motors. A PC running the MLCad Lego design system sends instructions to the printer via USB, and it then starts selecting bricks from its large-capacity feeder system.