For amputees, the struggle to connect with their bionic hands is a real challenge. But what if artificial intelligence could bridge this gap and make these hands feel more natural?
Meet Samoana Matagi, one of four participants in a groundbreaking study. In the photo, Matagi wears a bionic hand on one arm and a traditional body-powered hook on the other. The goal? To test a new approach that combines AI and special sensors to create a more intuitive prosthetic experience.
The Power of Shared Control
Marshall Trout, lead author of the study and a researcher at the University of Utah, explains the key concept: shared control. The bionic hand is designed to recognize the user's intent and then work together with them to complete tasks. When the sensors and AI are engaged, participants could reliably grasp a cup and take a sip. Without this assistance, however, they struggled, crushing or dropping the cup every time.
John Downey, an assistant professor at the University of Chicago, highlights the significance of this achievement. He points out that the ability to exert the right amount of grasp force is a major challenge in prosthetics. Many amputees become frustrated with their bionic hands and stop using them due to these difficulties.
The Evolution of Bionic Hands
Modern bionic hands are impressive, with motors that allow them to swivel, move individual fingers, and manipulate objects. They can even detect electrical signals from the muscles controlling these actions. However, as their capabilities have increased, so has the complexity of controlling them. Users must concentrate intently, which is not how a natural hand operates.
A natural hand, Trout explains, requires minimal cognitive effort for routine tasks. Specialized circuits in the brain and spine take over, allowing efficient and automatic completion of tasks. Our conscious mind only steps in when something unexpected happens.
Creating a Smart Prosthetic
Trout and his team aimed to develop a smart prosthetic that mimics this natural hand behavior. They wanted to recreate the intuitive squeeze and contact with a coffee cup, for example. To achieve this, they used AI and sensors to enable the bionic hand to share control with the brain.
The AI control system goes beyond simple muscle signal detection. It learns to recognize the intention behind the signal, such as the slightest twitch of a muscle that flexes the hand. This triggers the machine controller, which understands that the user is trying to grasp something.
To make this work, the scientists added proximity and pressure sensors to the bionic hand. These sensors allow the AI system to gauge distance and assess an object's shape. Meanwhile, pressure sensors on the fingertips provide feedback to the user, indicating how firmly the prosthetic hand is holding the object.
The Benefits of Shared Control
Jacob George, a professor at the University of Utah and director of the Utah NeuroRobotics Lab, emphasizes the importance of shared control. He notes that while robotic hands can outperform human users in certain tasks, people often don't like using them because they feel foreign and uncontrollable.
Downey explains that our connection to our own hands comes from the joint control of our thoughts and reflexes in the brain stem and spinal cord. Our brain doesn't need to micromanage every motion because of these subconscious reflexes.
The smart bionic hand developed by Trout and his team addresses this issue. As George puts it, "The machine is doing something, and the human is doing something, and we're combining those two together." This shared control is a crucial step towards creating prosthetic limbs that feel like a natural extension of the body.
The Future of Prosthetics
Even with the impressive advancements in bionic limbs, Downey emphasizes that human brains are still essential. He highlights the dynamic range of a natural hand, which can gently thread a needle and then firmly lift a child. This range is beyond what most robots can handle.
While bionic limbs continue to evolve and become more versatile, scientists believe that the desire for human control over artificial appendages will remain a constant. This shared control approach may be the key to creating prosthetics that feel truly integrated into the user's body.
And this is just the beginning. What other innovations might AI bring to the world of prosthetics? The possibilities are exciting, but there's still much to uncover. What do you think? Share your thoughts in the comments!
