Emerging Technologies

This AI-operated robotic hand moves with “unprecedented dexterity”

A 12-inch wafer is displayed at Taiwan Semiconductor Manufacturing Company (TSMC) in Xinchu January 9, 2007. TSMC, the world's top contract chip maker, posted an 18.4 percent fall in December sales on January 10, hurt by clients' unwanted stockpiles, but a recovery looms as a new crop of electronics devices hit store shelves. Picture taken January 9, 2007.  REUTERS/Richard Chung (TAIWAN) - RTR1L1ID

The Dactyl robotic hand was trained in a simulated environment. Image: REUTERS/Richard Chung

Kristin Houser
Writer, Futurism

Robotic high five

On Monday, researchers at OpenAI, the nonprofit AI research company co-founded by Elon Musk, introduced Dactyl, an AI system trained to control a robotic hand. According to the researchers, the system can manipulate physical objects in the hand with dexterity never before possible for AI.

The task Dactyl tackled might sound like something you’d teach a toddler: take this six-sided block and move it around until a certain side is on top. Unlike a toddler, though, Dactyl needed more than a century’s worth of experience to learn how to expertly complete the task. But thanks to powerful computers, the researchers were able to pack all that experience into just 50 “real-world” hours.

Image: OpenAI

Practice makes (almost) perfect

The researchers trained Dactyl in a simulated environment — that is, a digital setting with a computer-generated hand — using a technique called domain randomization. They built certain parameters into their simulated environment, such as the cube’s size or the angle of gravity, and then randomized those variables. They had multiple simulated hands doing this at once. By pushing Dactyl to adapt to so many different virtual scenarios, the researchers prepared the AI’s ability to adapt to scenarios in the real world.

After 50 hours of training in the simulated environment, the AI could manipulate a real-world robotic hand to successfully complete its given task 50 times in a row (a successful completion was one in which the system didn’t drop the block or take longer than 80 seconds). To figure out how to move the hand to complete the task, it simply needed to look at the block through a trio of cameras.

One algorithm to train them all

As the researchers note in their blog post, they trained Dactyl using the same algorithm that they used for OpenAI Five, a team of five neural networks trained to play the computer strategy game DOTA 2. Dactyl’s success proves it’s possible to build a general-purpose algorithm that can teach AI to complete two very different tasks. This could make it much easier for researchers to train AI for lots of different purposes in the future, since they wouldn’t need to start the process from scratch.

Loading...
Have you read?
Don't miss any update on this topic

Create a free account and access your personalized content collection with our latest publications and analyses.

Sign up for free

License and Republishing

World Economic Forum articles may be republished in accordance with the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International Public License, and in accordance with our Terms of Use.

The views expressed in this article are those of the author alone and not the World Economic Forum.

Stay up to date:

Emerging Technologies

Share:
The Big Picture
Explore and monitor how Biotechnology is affecting economies, industries and global issues
A hand holding a looking glass by a lake
Crowdsource Innovation
Get involved with our crowdsourced digital platform to deliver impact at scale
World Economic Forum logo
Global Agenda

The Agenda Weekly

A weekly update of the most important issues driving the global agenda

Subscribe today

You can unsubscribe at any time using the link in our emails. For more details, review our privacy policy.

3:35

These ‘underwater tractors’ are replanting seagrass and corals

Science once drove technology – but now the reverse is true. Here's how we can benefit

About us

Engage with us

  • Sign in
  • Partner with us
  • Become a member
  • Sign up for our press releases
  • Subscribe to our newsletters
  • Contact us

Quick links

Language editions

Privacy Policy & Terms of Service

Sitemap

© 2024 World Economic Forum