A robotic arm connected directly to a partially paralyzed man’s brain allows him to feed himself

A man with very limited upper body range of motion, unable to use his fingers for about 30 years, he used his head and smart robotic hands to eat desserts. (Image: Researcher)

Two robotic arms (a fork in one hand and a knife in the other) are placed on either side of a man sitting at a table with a cake on a plate. A computerized voice announces each action: “move fork to food” and “retract knife”. A paraplegic man subtly moves his left and right fists according to instructions such as “choose where to cut,” and the machine slices it into bite-sized pieces. Now: “Moving food to your mouth” is another subtle gesture of bringing a fork to your mouth.

In less than 90 seconds, a person with very limited upper body mobility who had been unable to use his fingers for about 30 years, used his head and a smart robotic hand to eat a dessert.

A team led by researchers from the Johns Hopkins Applied Physics Laboratory (APL) in Laurel, Maryland and the Department of Physical Medicine and Rehabilitation (PMR) at the Johns Hopkins Medical School has published a paper in the journal. Frontiers of Neurorobotics This describes this latest feat using a Brain Machine Interface (BMI) and a pair of modular prosthetic limbs.

Sometimes called the brain-computer interface, the BMI system provides a direct communication link between the brain and the computer. Computers decode nerve signals and “transform” them to perform a variety of external functions, from moving a cursor across the screen to now playing games. A bite of cake. In this particular experiment, muscle motor signals from the brain helped control the robotic prosthesis.

This research builds on more than 15 years of research in neuroscience, robotics, and software, originally led by APL in collaboration with the PMR Division as part of the Prosthetic Revolution Program sponsored by the US Defense Advanced Research Projects Agency. increase. (DARPA). A new paper outlines an innovative model of shared control that allows humans to operate a pair of robotic prosthetic limbs with minimal mental input.

“This shared control approach aims to leverage the unique capabilities of brain-machine interfaces and robotic systems, creating a ‘best of both worlds’ environment where users can personalize the behavior of smart prostheses,” said APL. Senior Project Manager in Research and Development, Senior Author of this paper, Tenore focuses on neural interfaces and applied neuroscience research.

“Our results are preliminary, but we are excited to give users with limited capabilities a sense of true control over increasingly intelligent assistive devices,” he added.

One of the most important advances in robotics presented in this paper is the combination of robot autonomy with limited human input, while allowing users to customize robot behavior to their liking. , the machine does most of the work.

“For robots to perform human-like tasks for people with reduced function, they would need human-like dexterity. should be controlled,” he explained. “Our goal is to give users easy control over the few things that are most important for a given task.”

Although the DARPA program officially ended in August 2020, teams at APL and Johns Hopkins School of Medicine, in collaboration with colleagues at other institutions, continue to demonstrate and explore the potential of this technology.

The next iteration of the system could integrate previous research that found that providing sensory stimulation to amputees could not only perceive a phantom limb, but also use muscle motor signals from the brain to control the prosthetic limb. There is a nature. The theory is that the addition of sensory feedback delivered directly to the human brain could help perform some tasks without the need for constant visual feedback in current experiments.

“This study is a great example of this philosophy of knowing that we had all the tools to demonstrate this complex bimanual activity in everyday life that people without disabilities take for granted.” said Tenore. “There are still many challenges ahead, such as improving task execution in terms of both accuracy and timing, and closed-loop control that does not require constant visual feedback.”


video A robotic arm connected directly to a partially paralyzed man’s brain allows him to feed himself

Back to top button