Mimicking the complex movement of a human hand is not an easy task for roboticists. As well as multiple bones and joints, there are tendons, ligaments, muscles, nerves and more to try and match. Researchers from the University of Cambridge have looked at simplifying the design and cutting back on the programming required to control a robotic hand, and developed a piano-playing soft skeleton bot that rocks the keyboard.
The robot hand was 3D-printed on a Stratasys Connex 5000 3D printer using soft and rigid materials that essentially replicated the bones and ligaments of a human hand. This was attached at the wrist to a UR5 robotic arm, which allowed for precise control of the skeletal hand using a Python API.
The fingers don't have any actuators or hydraulics so can't move on their own, but by twisting the hand at the wrist, the researchers managed to get digits to pick out keys on the piano keyboard and roll off simple tunes. And thanks to 3D printing, the researchers were able to vary the stiffness of the ligaments between the bones, allowing for playing and movement flexibility.
"Our bodies consist of smart mechanical designs such as bones, ligaments, and skins that help us behave intelligently even without active brain-led control," said project lead Dr. Fumiya Iida. "By using the state-of-the-art 3D printing technology to print human-like soft hands, we are now able to explore the importance of physical designs, in isolation from active control, which is impossible to do with human piano players as the brain cannot be 'switched off' like our robot."
The scientists looked at single finger tapping, thumb motion and hand span behavior. Four single fingers were used for the first series of experiments, where different control parameters were observed. The thumb motion experiments involved both the thumb and index finger of the design, aiming to replicate the behavior of human players when smoothly moving sequentially over notes. And jumps and transitions were the aim of the game when looking at full hand span behavior.
Force-sensing resistors on the piano fed playing data to an Arduino microcontroller in real time to measure behaviors and performance.
"We can use passivity to achieve a wide range of movement in robots: walking, swimming or flying, for example," said the team's Josie Hughes. "Smart mechanical design enables us to achieve the maximum range of movement with minimal control costs: we wanted to see just how much movement we could get with mechanics alone."
There are many examples of more advanced robot hands, and this one isn't going to win any piano playing competitions, but the researchers say that it could lead to the design of robots capable of more natural movement while cutting back on energy use.
The research could be broadened in the future to investigate more complex manipulation, potentially leading to the development of robots that use this approach to perform medical tasks or handle fragile objects more like we do.
A paper detailing the research has been published in the journal Science Robotics. The video below has more.
Source: University of Cambridge