Holding an object in your hand might seem easy, but it’s actually a very complicated and challenging task; if, for instance, an object starts to slip, you typically have a couple of milliseconds to react.
Scientists have been trying new approaches for improved control of robotic hands, particularly for use by amputees. A recent technology was able to combine individual finger control and automation for improved grasping and manipulation by successfully merging the fields of neuroengineering and robotics. This interdisciplinary approach was tested on three amputees and seven non-amputee subjects.
The neuroengineers achieved the intended finger movement from muscular activity on the amputee's stump, allowing for individual finger control of a prosthetic hand, which had never been done before. The robotics team enabled the robotic hand to take hold of objects and maintain contact with them for robust grasping. The amputee first performed a series of hand movements in order to train the algorithm through a machine learning paradigm. This taught the algorithm to decode user intention and translate it into finger movements of the prosthetic hand.
Concurrently, sensors placed on the amputee's stump detected muscular activity, which trained the algorithm to learn which hand movements corresponded to which patterns of muscular activity. Once the user's intended finger movements were acquired, this information could then be used to control individual fingers on the prosthetic hand. When the user tried to grasp an object, the robotic automation initiated. The algorithm told the prosthetic hand to close its fingers when an object was in contact with sensors on the hand’s surface.
This automatic grasping was designed to infer the shape of objects and grasp them based on tactile information alone, without any help of visual signals. The robotic hand has the ability to react within 400 milliseconds, and it is equipped with pressure sensors all along the fingers: it can react and stabilize the object before the brain can actually perceive that the object is slipping. While this promising technology can be used in in several neuro-prosthetic applications such as bionic hand prostheses and brain-to-machine interfaces, there are still many challenges remaining to implement this technology in a commercially available prosthetic hand for amputees. It is currently being tested and improved.