Johns Hopkins University engineers have developed a prosthetic hand capable of gripping and adjusting to objects with a level of dexterity comparable to the human hand. The device, described in Science Advances, incorporates a hybrid design that combines rigid and soft materials, allowing it to handle objects of various textures without damaging or mishandling them.
Unlike traditional robotic hands, which are typically either too rigid or too soft to replicate human touch, this prosthetic features a multifinger system with a flexible, rubberlike exterior and a rigid internal skeleton created through 3D printing. It is equipped with three layers of tactile sensors, inspired by the layers of human skin, enabling it to detect and respond to different shapes and textures. The hand is controlled by muscle signals from the forearm, and machine learning algorithms process sensory input to generate realistic touch sensations.
In laboratory tests, the prosthetic successfully identified and manipulated 15 different objects, including fragile items such as plastic cups filled with water and delicate stuffed toys. The device demonstrated a 99.69% accuracy rate in handling objects and automatically adjusted its grip to prevent slippage or excessive force.
The development team, led by biomedical engineer Sriramana Sankar, drew inspiration from the natural mechanics of the human hand, which combines both rigid and soft elements. The system incorporates electrical nerve stimulation to translate sensory feedback into signals recognizable by the brain, allowing users to experience a sense of touch.
Nitish Thakor, a professor of biomedical engineering at Johns Hopkins, emphasized that the innovation represents a step toward more advanced prostheses that integrate sensory perception. Future refinements could include stronger grip forces, additional sensors, and more durable materials. The research was supported by the U.S. Department of Defense and the National Science Foundation.
Photo credit: Sriramana Sankar/Johns Hopkins University
