OpenAI robot solves Rubik’s Cube one-handed

Joanna Estrada
October 17, 2019

AI research organization OpenAI have been hard at work building a general goal, self-learning robot with its robotics division Dactyl unveiling its humanoid robotic hand in 2018 - which is now being used to solve a Rubik's cube in less than 4 minutes flat.

Roboticists use simulation to reduce the amount of real-world training required for a robotic system to learn a task.

Dactyl stumbles, but eventually solves the Rubik's cube - with the team having a goal of seeing their AI-powered robotic appendages working on real-world tasks. The deal here isn't so much that it was able to solve the Rubik's cube, but that it did the task with a new level of dexterity.

Earlier this year, researchers at the University of California at Irvine unveiled an AI algorithm (sans robot arm) that can analyze more than 10 billion possible combinations to solve a Rubik's cube in just over a second. In comparison, the human record for solving a 3x3x3 cube single-handedly is now 6.82sec, held by American Max Park, 18. "The important difference between what they did there and what we're doing here is that those robots are very purpose-built", says Peter Welinder, research scientist and robotics lead at OpenAI.

While the other robots were built for one specific task, Dactyl is being developed as a self-learning robotic hand that has a near human approach to tasks solving.

The blink-and-you'll-miss-it clip shows a robot developed by two United States researchers cracking the tricky puzzle in a world-record time. It's being trained to replicate the instinctive use of hand mirroring that of a human child.

Other reports by Click Lancashire

Discuss This Article