The robot’s hand uses the “skin” to hold chopsticks, balls, and roll bubbles on the right | Science and technology news

[ad_1]

University of Cambridge researchers have created a 3D-printed hand that looks just as convincing when holding a few paper bubbles as it does a chopstick, ready for whatever noodle you throw at it.

By Tom Akers, Technology Reporter

Wed 12 Apr 2023 01:39, United Kingdom

From gripping a ball to gently grasping a chopstick, a new robotic hand developed by scientists in the UK can grip a range of objects by simply flicking its wrist and feeling its “skin”.

The 3D printed accessory was designed to be low cost and energy efficient, capable of performing complex movements despite not being able to use each finger independently.

Professor Fumiya Ida, of the Bioinspired Robotics Laboratory at the University of Cambridge, said the aim was to “simplify the hand as much as possible”.

Most advanced robots capable of similar feats to the human hand have fully robotic fingers, making them more difficult and expensive to produce.

But this cheaper alternative has proven itself remarkably well through more than 1,200 tests – including knowing how much pressure to apply to a particular object.

More science and tech news: The AI-generated newsreader debuts

Image: a robotic hand gently holding a chopstick. Pic: University of Cambridge

The “robot’s skin” helps judge the force required

While you should instinctively know how to gently handle the egg without smashing it and ruining your breakfast, the robots require training to recognize the correct amount of force required.

In this case, the researchers implanted the hand with sensors so it could sense what it was touching.

I used trial and error to see what types of grip would be successful—starting with balls and then moving on to everything from peach and bubble wrap to a computer mouse.

Study co-author Dr Thomas George Throthell, who is now at University College London, said the sensors were “kind of like the skin of a robot”.

He added, “We can’t say exactly what information the robot is getting, but it can theoretically estimate where the object was grabbed and how much force.”

The robot can also predict if it is going to drop something, and adapt accordingly.

This is a limited edition version of the story, so unfortunately this content is not available. Unlock the full version

The researchers hope that the robotic hand can be improved further, such as by adding computer vision capabilities and teaching it to exploit its surroundings to understand a wider range of objects.

The results are published in the journal Advanced Intelligent Systems.

[ad_2]