The engineers of the University of Johns Hopkins have developed a pioneer prosthetic hand that can grab luxury toys, bottles of water and other everyday objects such as a human, adjusting and carefully adjusting their understanding to avoid damaging or making bad management what it has.
The hybrid design of the system is the first for robotic hands, which have generally been too rigid or too soft to replicate the touch of a human by handling objects of different textures and materials. Innovation offers a promising solution for people with hands loss and could improve how robotic arms interact with their surroundings.
Details about the device appear today in Scientific advances.
“The objective from the beginning has been to create a prosthetic hand that we model based on the physical and detection of the human hand, a more natural prosthesis that works and feels like a lost limb,” said Sankar, a biomedical engineer from Johns Hopkins who directed the work. “We want to give people with loss of higher extremities the ability to interact safely and freely with their environment, feel and keep their loved ones without worrying about hurting them. “
The device, developed by the same neuroinger and laboratory of biomedical instruments that in 2018 created the first electronic “skin” of the world with a sense of human pain, presents a multifinger system with rubber polymers and an internal skeleton printed in rigid 3D. Its three layers of touch sensors, inspired by the layers of human skin, allow it to grab and distinguish objects in various surface shapes and textures, instead of just detecting touch. Each of its finger joints full of soft air can be controlled with the forearm muscles, and automatic learning algorithms focus the signals of artificial tactile receptors to create a sense of realistic touch, Sankar said. “The sensory information of its fingers translates to the language of the nerves to provide naturalistic sensory feedback through the stimulation of the electric nerve.”
In the laboratory, the hand identified and manipulated 15 everyday objects, including delicate stuffed toys, sponges of dishes and cardboard boxes, as well as pineapples, metal water bottles and other more resistant items. In the experiments, the device achieved the best performance compared to the alternatives, successfully managing objects with a accuracy of 99.69% and adjusting its grip as necessary to avoid mishaps. The best example was when they agilely picked up a thin and fragile plastic cup full of water, using only three fingers without abolling it.
“We are combining the strengths of rigid and soft robotics to imitate the human hand,” Sankar said. “The human hand is not completely rigid or purely soft: it is a hybrid system, with bones, soft joints and tissue that work together. That is what we want our prosthetic hand to achieve. This is a new territory for robotics and prostheses, which have not fully accepted this hybrid technology before.
To help amputates to recover the ability to feel objects while they are understood, the prostheses will need three key components: sensors to detect the environment, a system to translate those data into signals similar to the nerves and a way of stimulating the nerves so that the person can feel the sensation, said Nitit Thakor, a biometic engineering teacher at Johns Hopkins who directed the work.
Bioinspire technology allows the hand to work in this way, using muscle signs of the forearm, like most prostheses of the hand. These signs unite the brain and nerves, which allows the hand to flex, release or react based on their sense of touch. The result is a robotic hand that intuitively “knows” what is touched, just like the nervous system, said Thakor.
“If you are holding a cup of coffee, how do you know you are about to release it? Your palm and fingertips send signals to your brain that the cup is slipping,” Thakor said. “Our system is inspired neurally: models the tactile receptors of the hand to produce Nervelike messages so that the” brain “of the prostheses, or their computer, understand whether something is hot or cold, soft or hard or slipping from the grip.”
While research is an early advance for hybrid robotic technology that could transform prostheses and robotics, more work is needed to refine the system, Thakor said. Future improvements could include stronger grip forces, additional sensors and industrial grade materials.
“This hybrid skill is not only essential for next -generation prostheses,” Thakor said. “It is what the robotic hands of the future need because they will not only handle large and heavy objects. They will have to work with delicate materials such as glass, fabric or soft toys. That is why a hybrid robot, designed as the human hand, is so valuable: it combines soft and rigid structures, such as our skin, tissue and bone.”
Other authors include Wen-Yu Cheng from Florida Atlantic University; Jinghua Zhang, Ariel Slepyan, Mark M. Iskarous, Rebecca J. Greene, Renebander and Junjun Chen by Johns Hopkins; and Arnav Gupta of the University of Illinois Chicago.
This investigation was funded by the “Neuromorphic feedback subsidy: a strategy to improve prostheses and performance” of the Department of Defense through the Orthotics and Protection Results Research Program (W81xWH2010842) and the National Science Foundation.