Home / Blog / “Brain” controlled robotic arms

“Brain” controlled robotic arms

Patient with a C6 spinal cord injury connected to a computer reading data of brain activity captured by 6 implanted electrodes. Image credit: John Hopkins University

The quest towards Brain Computer Interfaces continues. At John Hopkins University researchers have implanted six electrodes in the brain, 3 in each hemisphere, of a paraplegic patient. The electrodes captures the electrical activity of the brain and send the data via an external gateway (see photo) to a computer for processing.

In a recently released report they show how the computer can convert the electrical activity of the brain into command to control two robotic arms (watch the video).

Is the computer able to “read” the patient mind? Not at all! What really happens is that the patient underwent a lengthy training process to learn how to generate electrical signals that can be interpreted in a correct way by the computer. The patient have been sitting in front of a computer screen showing simulated robotic arms and had to learn to move them by thinking “something”. Of course he was asked to think “move the arm to the right” and researchers tried to isolate some specific pattern that could be used by the computer as the command to move the robotic arm to the right. However, not necessarily the person had to think such a sentence. He could have though the sentence “violets are blue” and pretended that the electrical activity generated was to be used to move the robotic arm to the right. From the computer point of view it was exactly the same.  The training of the patient was necessary to have him learning to think something that would be easily recognised by the computer. By looking at the virtual robotic arm on the screen and seeing the resulting movement as he was thinking something over time he learnt what to think to get the intended result.

As you can see, we are quite far from having a computer reading our mind. It is more the other way around: the person has to learn what the computer could understand and “think” accordingly.

Nevertheless the result shown is impressive. That person was able to control 2 robotic arms nad have them cutting some food and bringing a morsel to his mouth. This requires the coordination of a quite complex set of movements, something that would have been impossible just two years ago. It is a huge progress in signal processing, not in artificial intelligence.

If you wonder about the six implanted electrodes, well, according to the researchers they have been able to find a material that is safe, does not generate inflammation and can keep working for five years, again, a significant progress. Some electrodes are implanted to capture the electrical activity (they are placed on the motor cortex), others are placed on the sensorial area of the brain and send pulses, based on the robotic arm movement to stimulate a sensation, thus making it easier for the patient to get a “feeling” on how his thoughts are affecting the robot (e.g. how the robotic arm is moving). This seems to speed up the training process. Again, notice how it is the patient that learns how to communicate with the computer.
I am not saying the on the computer side there is no learning whatsoever. Signal processing requires fine tuning and this is a sort of learning process.

About Roberto Saracco

Roberto Saracco fell in love with technology and its implications long time ago. His background is in math and computer science. Until April 2017 he led the EIT Digital Italian Node and then was head of the Industrial Doctoral School of EIT Digital up to September 2018. Previously, up to December 2011 he was the Director of the Telecom Italia Future Centre in Venice, looking at the interplay of technology evolution, economics and society. At the turn of the century he led a World Bank-Infodev project to stimulate entrepreneurship in Latin America. He is a senior member of IEEE where he leads the New Initiative Committee and co-chairs the Digital Reality Initiative. He is a member of the IEEE in 2050 Ad Hoc Committee. He teaches a Master course on Technology Forecasting and Market impact at the University of Trento. He has published over 100 papers in journals and magazines and 14 books.