Engineers from the University Carnegie Mellon in collaboration with experts from the University of Minnesota has made a breakthrough in the development of noninvasive methods of management of the robotic device. Using non-invasive brain-computer interface (BCI), scientists have created the world’s first robotic arm that is controlled by the human brain and have the ability to follow the direction of the mouse cursor on the computer screen. How notes portal Tech Explore reporting on the development, the use of a noninvasive method to control robotic devices will have a broad range of applications. For example, this technology can be very useful for paralyzed individuals.
Disadvantages of modern brain-computer interface
In the past, brain-computer interfaces showed very high control precision robotic objects, but for this purpose special implants that have violales in the human brain and track the desired signals. The introduction of these implants is difficult and quite dangerous task, requiring surgery. In addition, such devices are very expensive, but in fact may be very limited in their capabilities. Therefore, similar mechanisms will apply in very rare cases.
One of the main current problems of the sphere of the development of brain-computer interfaces is the creation of less invasive and better – completely non-invasive technology that will allow paralyzed people to control their environment or robotic limbs using the power of their thoughts than help millions of people worldwide to improve their quality of life.
The problem is that the use of non-invasive brain-computer interfaces that collect information on the brain signals via external electrodes, and not through implanted brain chips, accompanied by the transfer of signals along the brain amount of “noise” which reduces the control accuracy. So of noninvasive methods to visualize control robotic devices lose much technology with the use of implants. Despite this, the developers did not give up and try to establish more precise management practices that will not require surgery.
And, it seems, head of the Department of biomedical engineering, University of Carnegie Mellon, Professor Bina Heh succeeded.
“In the development of robotic devices controlled by thought via a special brain implants has achieved significant successes. It is really so. However, the key goal of such developments is the creation of non-invasive methods. Advances in neural decoding and the practical usefulness of non-invasive control of robotic arms will have a major impact on the development of non-invasive neurobiotic” — says He.
The world’s first controlled by thought, robotic hand
Using new methods of sensor technologies and machine learning-Hye along with his team was able to access deep brain signals to a high precision control raboroko. Thanks to non-invasive neuroimaging and the new paradigm of continuous persecution, the system learned to overcome noisy signals of EEG, which led to a significant improvement in neural decoding and facilitated continuous control robotic devices in real-time.
Using non-invasive BCI to control a robotic arm that tracks the cursor on the computer screen, Heh with the team for the first time demonstrated that the manipulator is now able to continuously follow the cursor. Previously, such devices follow the movement of the cursor sharp, discrete movements, as though trying to catch up to the team’s brain – now the limb follows the cursor smoothly and continuously.
In an article published in the journal Science Robotics, the team describes the creation of a new platform, which aims to improve “brain” and “computer” components of BCI by increasing user engagement and their learning as well as improvement of the spatial resolution of the data by non-invasive neural imaging EEG sources.
The article stated that a unique team approach to solving this problem have improved the training system for BCI almost 60 percent for the traditional tasks of centering and improved the quality of continuous monitoring of a computer cursor by more than 500 percent.
To date, the system was tested with the participation of 68 healthy people (up to 10 sessions for each), people (up to 10 sessions for each subject), including the virtual device and the robotic arm for continuous persecution. Clinical testing of the technology with the participation of real patients, the researchers plan to begin in the very near future.
Video of one of the participants of the experiment using the virtual cursor by using brain-computer interface, and robotic arm, controlled by thought for the task of continuous control of the randomly moving target
“Despite the technical problems with the use of non-invasive signals, we are fully committed to make this a safe and economical technology to people who can benefit from it,” says He.
“This work represents an important step in the development of non-invasive computer-brain interfaces – technology that someday may become a widespread subsidiary system to help people like smartphones”.
To discuss the news in our Telegram chat.