The artificial intelligence company, known as OpenAI, has trained a human-like robot hand to manipulate physical objects with unprecedented dexterity.
The robot is known as Dactyl was taught entirely in a simulation without any human input, and later transferred its knowledge to a real-world robot, eventually manipulating a six-sided cube. "Our results show that it’s possible to train agents in simulation and have them solve real-world tasks, without physically-accurate modelling of the world," the team at the Elon Musk co-founded company said.
Using a shadow dexterous hand, OpenAI placed a cube in the pal, of the hand and asked Dactyl to reposition it into a different orientation. Dactyl was taught how to solve the object reorientation task entirely in a simulation without any human interference.
"This gives us the best of both approaches: by learning in simulation, we can gather more experience quickly by scaling up, and by de-emphasizing realism, we can tackle problems that simulators can only model approximately."
Using humanoid hands to effectively manipulate objects has proved a long-standing challenge in robotic control. The completion of this project achieves a full cycle of artificial intelligence development that OpenAI has been working on for the past two years.
This shows that robot hands – and by extension, robots in general – can discover and adopt human behaviours to their own specific body types. Learn more about the OpenAI Dactyl hand by watching the video below.