Researchers have developed a software tool that could be used to control a robotic arm.
The technique could allow robotic arms to control robotic bodies.
The new software tool, which is described in a paper published in the journal Nature Nanotechnology, was created using data from the Human Connectome Project.
“It’s not that we invented the human brain, but rather that we’ve created the tools to understand how our brain works,” said lead author Zhenzhen Lu, a PhD student at the University of Toronto’s Department of Cognitive Sciences.
The tool, known as the “brain-machine interface” (BMI), is based on a neural network.
This network is a combination of thousands of neurons that represent a person, their environment and their cognitive function.
Researchers have been trying to develop a tool that can use this network to perform tasks that are thought to be relatively simple, like translating sentences.
But this type of brain-computer interface can be difficult to implement and requires special hardware and software.
In the new paper, Lu and colleagues developed a brain-interface platform that was both flexible and capable of controlling a robotic body.
The team created a neural-network-based platform that can be used by a human to control robots.
The system is built on a 3D-printer-compatible computer, a software framework, and a database of the human body.
This database includes an array of sensors and other data that are used to create the AI network, which can perform actions on the robot.
The researchers were able to control an arm that was capable of performing tasks like bending and lifting a person.
They also developed a neural interface that could interact with a robotic hand to perform simple movements like turning an arm on and off.
“We’re using a neural algorithm to translate between sentences and commands that the brain can’t do itself,” Lu said.
“We’re essentially using a machine to learn from the brain.”
In the future, the researchers envision using this AI-based system to control large-scale robotic arms that could lift or bend objects.
The study was funded by the U.S. Department of Energy and National Science Foundation.
The authors will present their findings at the upcoming IEEE International Conference on Robotics and Automation in September.
