Summary: New research shows that combining artificial intelligence with advanced height and pressure sensors allows a commercial bionic hand to grasp objects in a natural, intuitive way for amputees. By training a neural network in the grasping gut, each finger can independently “See” objects and automatically enter the right situation, improving accurate safety.
Participants performed everyday tasks such as lifting cups and picking up small objects with minimal cognitive strain and without much training. A system that shares a balanced control of the machine's purpose with the machine's assistance, which allows for a seamless, life-like use of the air hand.
Basic facts
- Environmental Management: AI-powered fingers use proximity and pressure sensors to create steerable, precise grasses.
- Cognitive load of the mind: Participants performed tasks with less mental effort and greater accuracy.
- Shared independence: The system has integrated user control with the help of AI to avoid collisions and increase vanity.
Source: University of Utah
Whether you're reaching for a mug, a pencil or someone's hand, you don't need to teach each of your fingers where needed to get the right grip.
The loss of that intrinsic ability is one of the many challenges for people with hand and hand problems. Even with the most advanced robotic varieties, these everyday objects come with an added mental load as users purposefully open and close their fingers around the target.
Researchers at the University of Utah are now using artificial intelligence to solve this problem. By integrating proximity and pressure sensors into a commercial bionic hand, and training a neural network in the grasping movement, the researchers developed an efficient, natural-like, intuitive way to grasp objects. When working in conjunction with artificial intelligence, study participants demonstrated greater safety, greater clarity and less mental effort.
Interestingly, the participants were able to perform many daily tasks, such as picking up small objects and lifting a cup, using different grip styles, all without extensive training or practice.
The research is led by Professor of Engineering Jacob A. George and Marshall Trout, a postdoctol researcher at the Utah Neurorobotic Lab, and appears Tuesday in the journal Environmental Communications.
“As life-like as bionic arms become, they're controlled at the moment and they're not easy or precise,” Trout said. “Approximately half of all users will abandon their prosthesis, often citing their poor controls and the burden of hearing loss.”
One problem is that most commercial bionic arms and hands don't have a way to replicate the normal gestures that give us precise, tactile gestures. Evil is not just a matter of emotional response, however. We also have subconscission models in our arrays that simulate and anticipate the interaction of a contributing object; A “Smart” hand would also need to learn these automatic responses over time.
Utah researchers tackled the first problem by outfitting a prosthetic hand, made by Haskata Prosthetics, with custom information. In addition to receiving pressure, these dogs were equipped with proximity sensors designed to replicate the positive experience. Fingers can see a weightless cotton ball dropped on them, for example.
In the second program, they train a neural network model on the proximity information so that the fingers will naturally move to a certain distance needed to hold the right part of the object. Because each finger has its own sensor and can see “in front of it, each digit works in harmony to create a perfect, stable match for everything.
But another problem remains. What if the user would intend to understand them in such a thing? What if, for example, they wanted to open their hand to throw something away? To address this last piece of the puzzle, researchers developed a bioinspired approach that involves sharing control between the user and the AI agent. The success of the method relied on finding the right balance between human and machine control.
“What we don't want is for the user to be fighting the control machine. On the contrary, here the machine improved the user's accuracy while simplifying tasks,” said Trout. “In fact, the machine realized their natural control to complete tasks without thinking about them.”
The investigators also conducted studies with four participants whose amputations occurred between the elbow and the wrist. In addition to advanced performance in standard tasks, they also try many daily tasks that require good motor control. Simple tasks, such as drinking from a plastic cup, can be very difficult for a victim; Soak it too soft and you'll throw it away, but rub too hard and you'll break it.
“By adding some artificial intelligence, we were able to load this feature into the prosthesis itself,” George said. “The end result is an intuitive and multi-tasking control, which allows simple tasks to become easy again.”
George solzbacher-chen professor in the college of engineering and engineering engineering and spencer fox cheating in the school's department of medicine and pharmaceuticals.
This project is part of the Utah Neurorobotic Lab's larger vision to improve the quality of life for amputees.
“The study team is also testing neareral junctions installed that allow people to control the prostheses with their mind and get a sense of touch coming back from this,” said George. “Next steps, the team plans to integrate this technology, so that their advanced sensors can improve the tactile function and the smart prosthesis can be seamlessly integrated with the control based on emotions.”
The course was published online at Natural Communication Under the title “Bionic Hand Sharing Human Machine Control Improves Grasp and Reduces Cognitive Load for Amputees.”
Coaunars include members of the neurorobotic lab Fredi Mino, Connor Olsen and Taylor Heramoto, as well as an Assistant Professor at the school's Department of Medicine in the Department of Biomedical Engineering.
Funding: Funding came from the National Institutes of Health and the National Science Foundation.
Important Questions Answered:
A: The system uses finger proximity and pressure sensors and a trained neural network to automatically position each finger for a stable, natural grip.
A: No. A collaborative framework that integrates people's goals with the help of a machine, preventing conflict and preserving user agency.
A: Current prostheses require a lot of cognitive effort; The new AI-driven system returns accurate, low-effort handling similar to natural manual labor.
Editing notes:
- This article was edited by the editor of neuroscience news.
- The journal is fully reviewed.
- Additional context added by our staff.
About this course Neurotech and Robotics News
Author: Evan lerner
Source: University of Utah
Contact: Evan Lerner – University of Utah
Image: This photo is posted in Neuroscience News
Actual research: Open access.
“Management of a shared human device with the Bionic Hand improves grip and reduces cognitive load for transradial amputees” by Jacob A. George et al. Natural Communication
-Catshangwa
Shared human machine management of a smart bionic hand improves grasping and reduces cognitive load for transradial amputees
Bionic hands can replicate many of the movements of the human hand, but our ability to control these bionic objects is limited. The evil of the Human'Sun books is partly due to control loops driven by emotional feedback.
Here, we describe the integration of proximity and pressure sensors into a commercial advantage to enable autonomy and demonstrate that continuous collaboration between the autonomous hand and the user improves intelligence and user experience.
Artificial Intelligence moved each finger up to the point of contact while the user was intuitively controlled by the surface ectromlomblomblography. It is bioinspired with a limited weight of the combined machine and the purpose of the user. Shared control led to greater security, greater accuracy, and less mental burden.
Demonstrations include Interact and axteee participants using a modified prosthesis to perform real-world tasks with different grip patterns. Therefore, providing autonomy to some bionic hands presents a flexible and general way of controlling more and more accurately.
