CMU researchers have successfully developed the first-ever mind-controlled robotic arm without invasive brain implants.

Carnegie Mellon University’s researchers, in cooperation with the University of Minnesota, have made a major step in inventing robotic limbs that can be controlled solely by our minds. Researchers use “noninvasive brain-computer interface (BCI)” to develop a mind-controlled robotic arm that shows the ability to continuously track and follow a computer cursor.

The aims of this project is to develop robotic devices that can be controlled by thoughts without using invasive brain implants.

Formerly, BCIs need the signals sensed from brain implants in order to control robotic devices with high precision, allowing patients who suffer from paralysis and movement disorders to perform many daily tasks more effectively. However, it needs a massive amount of medical and surgical expertise to install brain implants. Factors such as cost and the potential risks to subjects must be considered as well. That’s why only few clinical cases were able to use brain implants.

As a result, BCI research is focusing on developing less invasive or “noninvasive” technology. Thus, if it is successful, it will benefit the lives of those who need it the most. Unfortunately, initial results of BCIs that use noninvasive external sensing indicate ‘negative’ signals, lower resolution and less precise control. However, that did not make BCI researchers give up because the reward for overcoming this challenge is so precious for our future.
‍
Here, Bin He, department head and professor of biomedical engineering at CMU, is reaching the finish line of this challenge.
‍

He says, “There have been major advances in mind controlled robotic devices using brain implants. It’s excellent science, but noninvasive is the ultimate goal. Advances in neural decoding and the practical utility of noninvasive robotic arm control will have major implications on the eventual development of noninvasive neurorobotics.”

‍

“Novel sensing and machine learning techniques are the keys”

He and his lab use “novel continuous pursuit paradigm” and “noninvasive neuroimaging” to access the signals in the brain, improving ability to control a robotic arm. In the same way, with these technologies, He can overcome the noisy EEG signals leading to considerably improve EEG-based neural decoding, and facilitate real-time continuous 2D robotic device control. 

Now, for the first time ever, the results of using a noninvasive BCI which tracks a cursor on a computer screen indicated a good sign; the robotic arm is able to follow the cursor continuously and smoothly.

The technology which can be used by just thoughts alone will no longer exist only in sci-fi movies. 

He says, “Despite technical challenges using noninvasive signals, we are fully committed to bringing this safe and economic technology to people who can benefit from it. This work represents an important step in noninvasive brain-computer interfaces, a technology which someday may become a pervasive assistive technology aiding everyone, like smartphones.”
‍

‍

Retrieved from: https://engineering.cmu.edu/news-events/news/2019/06/20-he-sci-robotics.html