Researchers at the Technical University of Munich (TUM) have developed a soft, pneumatic hand exoskeleton that can help people with paralysis regain the ability to grasp objects. Combining intelligent control with a low-cost design, the so-called Soft Hand Exoskeleton could enable people with neurological disorders or injuries to regain greater independence in their daily lives.
The „Soft-Hand-Exosceleton“is equipped with air cushions that are inflated through 13 tubes. This allows individual fingers to bend and extend while also supporting wrist movement. As a result, the system assists precisely those hand movements required to securely grasp and hold everyday objects such as a glass, a plate or cutlery.
To ensure the fabric glove responds at the right moment, sensors measure muscle activity in the forearm. Using machine learning, the system can reliably detect when a person intends to grasp an object. Additional motion sensors ensure that the grip remains closed while the object is being carried. “To prevent an object from being accidentally dropped, we use additional motion sensors to detect transport movements and reliably keep the exoskeleton’s grip closed during these movements,” explains researcher Nicolas Berberich.
According to the researchers, the system detects the user’s intention to grasp with an accuracy of 97 percent. At the same time, the glove stands out for its simple and inexpensive design. “Our solution is intelligent in two ways,” explains Dr. John Nassour. “On the one hand, we developed a reliable method for predicting grasping intentions from muscle signals with 97 percent accuracy. On the other hand, we created hardware that provides optimal support for the intended movements.”
A key factor in the development of the system was the collaboration with a patient suffering from amyotrophic lateral sclerosis (ALS). Although he had almost completely lost control of his hands and was only able to move the first joint of his thumb, the system correctly recognized his intention to grasp in nine out of ten cases. Using the exoskeleton, he was able to hold a fork for the first time in four years and successfully pick up and place small cubes. According to the researchers, just five minutes of training with a specially designed video game significantly improved his ability to control the system.
“This patient, who is affected by one of the most severe neurological diseases, has shown us that our Soft Hand Exoskeleton can provide meaningful support,” says Prof. Gordon Cheng, Head of the Institute for Cognitive Systems. “We are now adapting the concept for other patient groups, such as stroke survivors.”
Neurologist Prof. Tobias Wächter from the project partner Klinik Passauer Wolf also sees considerable potential in the technology: “In principle, this glove can help all people suffering from flaccid paralysis,” including patients with peripheral nerve injuries following motorcycle or bicycle accidents as well as people with polyneuropathy.