Share This Post

Featured News / Main Slider

Mind-Controlled Robotic Arm Developed Without the Need for Brain Implants

Mind-Controlled Robotic Arm Developed Without the Need for Brain Implants

A team of scientists from Carnegie Mellon University, in partnership with the University of Minnesota, have made a remarkable discovery in the field of noninvasive robotic device control. Scientists using a noninvasive brain-computer interface (BCI) have built the first-ever successful mind-controlled robotic arm displaying the ability to constantly track and follow a computer cursor.

Being able to noninvasively control robotic devices using just thoughts will have wide-ranging applications, especially helping the lives of paralyzed patients and those having movement disorders.

BCIs have been proven to accomplish good performance for regulating robotic devices using just the signals detected from brain implants. When robotic devices can be regulated with high precision, they can be used to perform a range of everyday tasks. So far, however, BCIs successful in regulating robotic arms have employed invasive brain implants. These implants necessitate a considerable amount of surgical and medical proficiency to properly install and operate, not to mention the cost and possible hazards to subjects, and as such, their use has been restricted to just a few clinical cases.

A huge challenge in BCI research is to create less invasive or even totally noninvasive technology that would enable paralyzed patients to regulate their robotic limbs or environment using their own “thoughts.” That kind of noninvasive BCI technology, if effective, would bring the much needed technology to many patients and even possibly to the general population.

However, BCIs that employ noninvasive external sensing, instead of brain implants, receive “dirtier” signals, leading to the present lower resolution and less precise control. Therefore, when using only the brain to regulate a robotic arm, a noninvasive BCI is unable to stand up to using implanted devices. In spite of this, BCI scientists have marched forward, their aim fixed on the prize of a less- or non-invasive technology that could aid patients globally on an everyday basis.

Bin He, Trustee Professor and Department Head of Biomedical Engineering at Carnegie Mellon University, is realizing that goal, one crucial discovery at a time.

“There have been major advances in mind-controlled robotic devices using brain implants. It’s excellent science,” says He. “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.”

Using new sensing and machine learning methods, He and his lab have been able to reach signals deep inside the brain, accomplishing a high resolution of control over a robotic arm. With noninvasive neuroimaging and a novel nonstop pursuit paradigm, He is overcoming the noisy EEG signals resulting in considerably enhanced EEG-based neural decoding, and enabling real-time uninterrupted 2D robotic device control.

With the help of a noninvasive BCI to regulate a robotic arm that is tracking a cursor on a computer screen, for the first time ever, He has demonstrated in human subjects that a robotic arm can currently follow the cursor uninterruptedly. However, robotic arms regulated by humans noninvasively had earlier followed a moving cursor in irregular, discrete motions — as though the robotic arm was attempting to “catch up” to the brain’s commands — now, the arm follows the cursor in a smooth, unbroken path.

The team defined a new framework that addresses and enhances upon the “brain” and “computer” components of BCI by expanding user engagement and training, as well as spatial resolution of noninvasive neural data via EEG source imaging. Details of the work can be found in Science Robotics.

The paper, “Noninvasive neuroimaging enhances continuous neural tracking for robotic device control,” outlines that the team’s unique approach to solving this issue improved BCI learning by almost 60% for traditional center-out operations, it also improved continuous tracking of a computer cursor by more than 500%.

The technology also has uses that could help a range of people, by providing safe, noninvasive “mind control” of devices that can permit people to interact with and regulate their environments. The technology has, so far, been tested in 68 able-bodied human participants (up to 10 sessions for each subject), including virtual device control and manipulating of a robotic arm for uninterrupted pursuit. The technology is directly valid for patients, and the team hopes to carry out clinical trials in the months ahead.

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.

Bin He, Trustee Professor and Department Head of Biomedical Engineering, Carnegie Mellon University

AzoRobotics

Share This Post

Leave a Reply