Skip to main content

OpenAI’s robotic hand doesn’t need humans to teach it human behaviors

Gripping something with your hand is one of the first things you learn to do as an infant, but it’s far from a simple task, and only gets more complex and variable as you grow up. This complexity makes it difficult for machines to teach themselves to do, but researchers at Elon Musk and Sam Altman-backed OpenAI have created a system that not only holds and manipulates objects much like a human does, but developed these behaviors all on its own.

Many robots and robotic hands are already proficient at certain grips or movements — a robot in a factory can wield a bolt gun even more dexterously than a person. But the software that lets that robot do that task so well is likely to be hand-written and extremely specific to the application. You couldn’t for example, give it a pencil and ask it to write. Even something on the same production line, like welding, would require a whole new system.

Yet for a human, picking up an apple isn’t so different from pickup up a cup. There are differences, but our brains automatically fill in the gaps and we can improvise a new grip, hold an unfamiliar object securely, and so on. This is one area where robots lag severely behind their human models. And furthermore, you can’t just train a bot to do what a human does — you’d have to provide millions of examples to adequately show what a human would do with thousands of given objects.

The solution, OpenAI’s researchers felt, was not to use human data at all. Instead, they let the computer try and fail over and over in a simulation, slowly learning how to move its fingers so that the object in its grasp moves as desired.

The system, which they call Dactyl, was provided only with the positions of its fingers and three camera views of the object in-hand — but remember, when it was being trained, all this data is simulated, taking place in a virtual environment. There, the computer doesn’t have to work in real time — it can try a thousand different ways of gripping an object in a few seconds, analyzing the results and feeding that data forward into the next try. (The hand itself is a Shadow Dexterous Hand, which is also more complex than most robotic hands.)

In addition to different objects and poses the system needed to learn, there were other randomized parameters, like the amount of friction the fingertips had, the colors and lighting of the scene, and more. You can’t simulate every aspect of reality (yet), but you can make sure that your system doesn’t only work in a blue room, on cubes with special markings on them.

They threw a lot of power at the problem: 6144 CPUs and 8 GPUs, “collecting about one hundred years of experience in 50 hours.” And then they put the system to work in the real world for the first time — and it demonstrated some surprisingly human-like behaviors.

The things we do with our hands without even noticing, like turning an apple around to check for bruises or passing a mug of coffee to a friend, use lots of tiny tricks to stabilize or move the object. Dactyl recreated several of them, for example holding the object with a thumb and single finger while using the rest to spin to the desired orientation.

What’s great about this system is not just the naturalness of its movements and that they were arrived at independently by trial and error, but that it isn’t tied to any particular shape or type of object. Just like a human, Dactyl can grip and manipulate just about anything you put in its hand, within reason of course.

This flexibility is called generalization, and it’s important for robots that must interact with the real world. It’s impossible to hand-code separate behaviors for every object and situation in the world, but a robot that can adapt and fill in the gaps while relying on a set of core understandings can get by.

As with OpenAI’s other work, the paper describing the results is freely available, as are some of the tools they used to create and test Dactyl.



from Gadgets – TechCrunch https://ift.tt/2mTTVqD

Comments

Popular posts from this blog

Mother's Day 2020 gift ideas: 18 gadgets and gizmos for tech-savvy Aussie mums

Raising a family is not an easy job, and the women who care for us each and every day deserve to be told how special they are each and every day. While we tend to forget to do that, Mother’s Day reminds us we need to celebrate the women in our lives, whether they’re our own mothers or our wives and partners helping us raise the young ones. Mother’s Day 2020 is fast approaching (with under two weeks to go), and there’s a pretty good chance you won’t be able to take her out to her favourite restaurant this year, or even get to a store to shop for something she might like. So we have to get creative, and TechRadar’s Australian team has put together this little list of great tech gift ideas that you can buy online and have delivered in time for May 10. But you will need to get a wriggle on as delivery supply chains are under strain with more people shopping online. Whether she’s a whiz in the kitchen, loves to cosy up with a book or entertain at home, we’ve got a gadget or gizmo that’s s...

Amazon Australia has specials on Bose products all this week

December may have just begun, but the world's largest online marketplace is already feeling the Christmas spirit.  To kick off the month’s festivities, Amazon Australia is celebrating  ‘7 Days of Deals’ with Bose's superb audio hardware discounted each day. To begin with, the very popular (and rightly so) Bose QuietComfort 35 II and the more expensive Bose Noise Cancelling Headphones 700 are available for less. To sweeten the deal, Amazon will throw in an Echo speaker as a bonus as well. When you purchase the superb Bose Headphones 700, you will receive a free Amazon Echo Show 5, or if you’d prefer the Bose QuietComfort 35 II, you’ll receive a complimentary Echo Dot. The offer is valid until December 8, or while stocks last. You can buy the same bundles, for the same price if you make the purchase via the Echo Dot or the Echo Show 5 product pages on Amazon. Just make sure you select the bundled headphone in the 'add other items' section on the right. Best noi...

Valentine's Day flowers: the best online flower delivery services

February 14 will be here before you know it, and if you, like many others, are searching for that perfect gift, then placing an online order for Valentine's Day flowers is always an easy and romantic option. You can order a beautiful floral arrangement in minutes from a variety of online retailers, including; 1-800-Flowers, Amazon, ProFlowers, Teleflora, and many more. To help you sort through all the Valentine's Day offers, we've rounded up the best online flower delivery services in both the USA and the UK and listed their current promotions. We've also included delivery charges and made sure to mention if you can allocate specific days for delivery. There's a fantastic range of bouquets and gifts available from our selection of florists below, and online delivery from a specialist means you don't have to worry about the usual hassle of buying from a store and getting them home safely. We'll be updating this page as we get closer to the big day so you...