Cambridge University scientists create robotic hand able to hold objects

    The 3D-printed hand was initially trained using plastic balls

Scientists have designed a robotic hand that can grasp and hold objects using only the movement of its wrist. 

A group at the University of Cambridge created the 3D-printed hand.

It was put through over 1,200 tests using things like a peach, a computer mouse, and bubble wrap with sensors that allowed it to "sense" what it was touching.

According to researchers, it was extremely difficult to replicate the capabilities of the human hand because it was so intricate.

Dr. Thomas George-Thuruthel, who previously worked at the University of Cambridge but now teaches robotics and artificial intelligence at University College London, stated: The pressure that is being applied to the object is measured by the sensors, which are similar to the skin of the robot.

"The robot can theoretically estimate where the object has been grasped and with how much force," although "we can't say exactly what information the robot is getting."

Because it did not require its fingers to move independently, the technology, according to the researchers, was inexpensive and energy efficient.

    This hand was only capable of passive, wrist-based movement and the individual fingers were not fully motorised

According to the team, humans instinctively know how much force to apply when picking up an egg, but this was difficult for a robot.

The egg could break if the robot applied too much force, or it could fall if it applied insufficient pressure.

On the 14 items it was tested on, the robotic hand was able to successfully grasp 11 of them.

Fumiya Iida, teacher of advanced mechanics at the College of Cambridge's branch of designing, said: " The wide range of motion that can be achieved without the use of any actuators is the main benefit of this design.

"We want the hand to be as simple as possible.

"We can get heaps of good data and a serious level of control with practically no actuators, so that when we truly do add them, we'll get more perplexing conduct in a more proficient bundle."