Sometimes ten fingers just aren't enough.
Scientists have created a wrist-mounted robot with an extra two fingers that let you outperform your average hands. The robot is attached to a glove with motion sensors that trace your actual fingers' movements and pose the extra fingers accordingly.
The robotic fingers were developed by MIT researchers that hope the device can one day help elderly people and people with disabilities. The two fingers extend next to the thumb and pinky and allow the wearer to do routine tasks much more quickly and efficiently.
"This is a completely intuitive and natural way to move your robotic fingers," says Harry Asada, professor of engineering at MIT. "You do not need to command the robot, but simply move your fingers naturally. Then the robotic fingers react and assist your fingers."
Asada believes that with training, users may begin to feel as if the extra fingers are truly part of their body. People with disabilities could use the device for tasks such as opening jars or envelopes and lifting heavy things.
Asada and graduate student Faye Wu presented the device at the Robotics: Science and Systems conference in Berkeley. Algorithms that Asada and Wu developed after studying the precise physiological details of hand gestures control the "supernumerary robotic fingers".
Their studied hand gestures to figure out the underlying patterns in the fingers' movements. They found that two general patterns of motion are used over and over when grasping objects--bringing fingers together and then twisting them. Breaking the motions down into these patterns, they then tested to see if such patterns would be maintained with seven fingers.
Attaching sensors to a glove, Wu used the glove to pick up objects. While holding the objects, she positioned the robotic fingers to support the grasp and measured and recorded the angles at which the fingers carried the weight. Then, using this information, the researchers created the algorithm to combine the movements of the natural fingers with those of the robotic ones to form the patterns seen when grasping.
"Right now we're looking at posture, but it's not the whole story," Wu says. "There are other things that make a good, stable grasp. With an object that looks small but is heavy, or is slippery, the posture would be the same, but the force would be different, so how would it adapt to that? That's the next thing we'll look at."
They also acknowledge that grasping patterns and other gestures likely differ from person to person. Wu hopes to one day create a human-robotic gesture pattern library that will help the robotic fingers adapt to each person's preference.
This development could change the way humans and robots interact.
