A major breakthrough that allows people to control a robotic arm using only their minds has been made by researchers at the University of Minnesota. The research could help millions of people who suffer from neurodegenerative diseases, or are paralyzed.
Bin He, lead researcher on the study and a University of Minnesota biomedical engineering professor said that people were able to move the robotic arm just by imagining moving their arms. He added that this is the first time in the world that people can use only their thoughts, without a brain implant, to operate a robotic arm to grasp and reach objects in a complex 3D environment.
The technique used is noninvasive and is called the electroencephalography (EEG) based brain computer interface. Weak electrical activity of the subjects’ brain is recorded through a specialized, high tech EEG cap fitted with 64 electrodes. The “thoughts” are then converted into action by advanced machine learning and signal processing.
Eight healthy human subjects wearing the EEG cap were used for experimental sessions of the study. Participants gradually learned how to control a robotic arm in 3D space by imagining moving their own arms without actually moving them. After having learnt to control a virtual cursor on computer screen, they then progressed and learned to control a robotic arm to grasp and reach objects at specific locations on a table. They were eventually able to move the robotic arm to reach and grasp objects in random locations on a table and then move these from the table to a shelf with three layers. This was achieved by only thinking about these movements.
The average success rate of all eight subjects was above 80 percent when they controlled a robotic arm to pick up objects in fixed locations, while a success rate of above 70 percent was achieved when they moved objects from the table onto the shelf.
He is excited by all subjects performing the tasks using a completely noninvasive technique. The team sees huge potential for this research to help people who suffer from neurodegenerative diseases, or are paralyzed, as they could become more independent without the need for a surgical implant.
According to the researchers, the brain computer interface technology works due to the geography of the area of the cerebrum that governs movement – the motor cortex. When we move, or think about moving, the neurons in the motor cortex produce miniscule electric currents. Thinking about a different movement activates a different group of neurons, an occurrence that was confirmed by cross-validation in He’s previous study. In that study, He used using functional MRI. He noted that the brain computer interface used by the University of Minnesota researchers was made possible by using advanced signal processing to distinguish between the different groups of neurons activated by various movements.
The robotic arm research is a continuation of He’s research that was published three years ago. In the previous study, subjects were able to fly a small quadcopter using the noninvasive EEG technology.
He is ecstatic with the high success rate in a group of people, as they weren’t at all sure that moving a more complex robotic arm to grasp and move objects using this brain-computer interface technology could even be achieved three years ago.
He expects the next step of his research will be to expand on the development of the brain-computer interface technology.
This could involve attaching a brain controlled robotic prosthetic limb to a person’s body, or investigate how this technology could work with someone who has suffered from a stroke, or is paralyzed.
The study was published in Scientific Reports.