Low-cost, energy-efficient robotic hand could help us grasp the future

Cambridge University researchers developed a novel robotic hand that works with minimal finger actuation.
Kavita Verma
A 3D printed robotic hand with tactile sensors
A 3D printed robotic hand with tactile sensors, designed for energy-efficient grasping, holds a delicate object without exerting excessive force.

University of Cambridge 

In a significant breakthrough, researchers at the University of Cambridge have designed an energy-efficient robotic hand that can grasp a variety of objects with minimal finger actuation, according to a study published on April 11 in Advanced Intelligent Systems.

By relying on passive wrist movement and tactile sensors embedded in its 'skin,' the 3D-printed hand can carry out complex movements, paving the way for low-cost, energy-efficient robotics with more natural and adaptable activities.

Grasping objects of varying sizes, shapes, and textures poses a significant challenge for robots, but Cambridge researchers have overcome this issue by focusing on passive movement. According to them, the robot hand is easy to control and consumes significantly less energy than fully motorized robotic hands.

This robotic hand is expected to improve robotic grasping

Professor Fumiya Iida's Bio-Inspired Robotics Laboratory at Cambridge's Department of Engineering has been developing potential solutions to improve robotic grasping. Dr. Thomas George-Thuruthel, a co-author of the study, said, "In earlier experiments, our lab has shown that it's possible to get a significant range of motion in a robot hand just by moving the wrist."

The 3D-printed anthropomorphic hand was equipped with tactile sensors to sense what it was touching. The researchers conducted over 1,200 tests to evaluate the hand's ability to grasp objects without dropping them. Initially trained using 3D-printed plastic balls, the hand successfully held 11 out of 14 different objects, including a peach, a computer mouse, and a roll of bubble wrap.

Dr. Kieran Gilday, the study's first author, explained in a press release, "The tactile sensors give the robot a sense of how well the grip is going, so it knows when it's starting to slip. This helps it to predict when things will fail."

"The sensors, which are sort of like the robot's skin, measure the pressure being applied to the object," said George-Thuruthel. "We can't say exactly what information the robot is getting, but it can theoretically estimate where the object has been grasped and with how much force."

The passive design of the Cambridge-developed robotic hand, combined with a small number of sensors, streamlines control, provides a wide range of motion, and simplifies the learning process.

Future developments could include integrating computer vision capabilities and teaching the robot to exploit its environment, enabling it to grasp an even broader range of objects.

The team is now working to improve the robot hand's accuracy and speed and exploring new applications for the technology. They are also developing a more advanced version of the hand that can sense and respond to changes in its environment, such as temperature or humidity.

The complete study was published in Advanced Intelligent Systems on April 11 and can be found here.

Study abstract:

Scientists have developed a minimal-expense, energy-productive mechanical hand that can grasp various items - and not drop them - utilizing only the development of its wrist and the inclination in its 'skin.' The key innovation of this approach lies in the use of an adaptive wrist mechanism that allows the hand to automatically adjust its grip force and orientation based on the ball's movements.

The researchers developed a prototype of the robot hand and conducted a series of experiments to evaluate its performance. The experiments involved training the robot to catch a ball thrown at various speeds and angles and testing its ability to maintain a firm grip on the ball while performing different tasks.

The results of the experiments showed that the adaptive wrist mechanism significantly improved the robot's performance in terms of energy efficiency and success rate. The robot hand learned how to adjust its grip force and orientation to catch the ball without dropping it, even when faced with unexpected variations in the ball's trajectory.

Add Interesting Engineering to your Google News feed.
Add Interesting Engineering to your Google News feed.
message circleSHOW COMMENT (1)chevron
Job Board